Commit
·
44b2580
1
Parent(s):
fd317ec
init commit
Browse files- README.md +14 -0
- attn_gate_weights.pth +3 -0
- config.json +33 -0
- trainer_state.json +3542 -0
README.md
CHANGED
|
@@ -1,3 +1,17 @@
|
|
| 1 |
---
|
| 2 |
license: mit
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: mit
|
| 3 |
+
library_name: transformers
|
| 4 |
+
base_model:
|
| 5 |
+
- deepseek-ai/DeepSeek-R1-Distill-Qwen-32B
|
| 6 |
+
base_model_relation: adapter
|
| 7 |
---
|
| 8 |
+
|
| 9 |
+
|
| 10 |
+
## SeerAttention-DeepSeek-R1-Distill-Qwen-32B-AttnGates
|
| 11 |
+
|
| 12 |
+
This repo only contains the AttnGates' weights for deepseek-ai/DeepSeek-R1-Distill-Qwen-32B.
|
| 13 |
+
|
| 14 |
+
[SeerAttention](https://arxiv.org/abs/2410.13276) introduces learnable AttnGate modules to accelerate the computationally intensive prefill stage of long-context large language models (LLMs) via dynamic block-level sparsity. The AttnGates are trained in a parameter-efficient self-distillation framework, where they learn to mimic the 2D max-pooled attention patterns of the original frozen model, preserving its integrity while avoiding costly retraining. During inference, these gates generate block-sparse binary masks by applying threshold/TopK to their learned soft scores, enabling efficient computation through a custom block-sparse FlashAttention kernel.
|
| 15 |
+
|
| 16 |
+
Original Github Repo
|
| 17 |
+
https://github.com/microsoft/SeerAttention.
|
attn_gate_weights.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:91a77643b3dfdcc14e50160f1f2ba6bd493fd456babd8d02df06d9991d5a9ab3
|
| 3 |
+
size 251668686
|
config.json
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"Qwen2ForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"base_model": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
|
| 8 |
+
"bos_token_id": 151643,
|
| 9 |
+
"eos_token_id": 151643,
|
| 10 |
+
"hidden_act": "silu",
|
| 11 |
+
"hidden_size": 5120,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 27648,
|
| 14 |
+
"max_position_embeddings": 131072,
|
| 15 |
+
"max_window_layers": 64,
|
| 16 |
+
"model_type": "qwen2",
|
| 17 |
+
"num_attention_heads": 40,
|
| 18 |
+
"num_hidden_layers": 64,
|
| 19 |
+
"num_key_value_heads": 8,
|
| 20 |
+
"rms_norm_eps": 1e-05,
|
| 21 |
+
"rope_theta": 1000000.0,
|
| 22 |
+
"seerattn_gate_block_size": 64,
|
| 23 |
+
"seerattn_gate_force_double": false,
|
| 24 |
+
"seerattn_gate_hidden_size": 128,
|
| 25 |
+
"seerattn_gate_type": "Qavg_Kmaxminavg",
|
| 26 |
+
"sliding_window": null,
|
| 27 |
+
"tie_word_embeddings": false,
|
| 28 |
+
"torch_dtype": "bfloat16",
|
| 29 |
+
"transformers_version": "4.44.2",
|
| 30 |
+
"use_cache": true,
|
| 31 |
+
"use_sliding_window": false,
|
| 32 |
+
"vocab_size": 152064
|
| 33 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,3542 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.4284490145672665,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 500,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.000856898029134533,
|
| 13 |
+
"grad_norm": 0.2508656084537506,
|
| 14 |
+
"learning_rate": 5e-05,
|
| 15 |
+
"loss": 2.5931,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.001713796058269066,
|
| 20 |
+
"grad_norm": 0.24156232178211212,
|
| 21 |
+
"learning_rate": 0.0001,
|
| 22 |
+
"loss": 2.6137,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.002570694087403599,
|
| 27 |
+
"grad_norm": 0.25357234477996826,
|
| 28 |
+
"learning_rate": 0.00015,
|
| 29 |
+
"loss": 2.5697,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.003427592116538132,
|
| 34 |
+
"grad_norm": 0.23629046976566315,
|
| 35 |
+
"learning_rate": 0.0002,
|
| 36 |
+
"loss": 2.3679,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.004284490145672665,
|
| 41 |
+
"grad_norm": 0.23719146847724915,
|
| 42 |
+
"learning_rate": 0.00025,
|
| 43 |
+
"loss": 2.3205,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.005141388174807198,
|
| 48 |
+
"grad_norm": 0.22218209505081177,
|
| 49 |
+
"learning_rate": 0.0003,
|
| 50 |
+
"loss": 2.1263,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.005998286203941731,
|
| 55 |
+
"grad_norm": 0.20407630503177643,
|
| 56 |
+
"learning_rate": 0.00035,
|
| 57 |
+
"loss": 1.9481,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.006855184233076264,
|
| 62 |
+
"grad_norm": 0.1823188215494156,
|
| 63 |
+
"learning_rate": 0.0004,
|
| 64 |
+
"loss": 1.7466,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.007712082262210797,
|
| 69 |
+
"grad_norm": 0.15546730160713196,
|
| 70 |
+
"learning_rate": 0.00045000000000000004,
|
| 71 |
+
"loss": 1.5499,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.00856898029134533,
|
| 76 |
+
"grad_norm": 0.14485396444797516,
|
| 77 |
+
"learning_rate": 0.0005,
|
| 78 |
+
"loss": 1.4354,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.009425878320479864,
|
| 83 |
+
"grad_norm": 0.16174261271953583,
|
| 84 |
+
"learning_rate": 0.00055,
|
| 85 |
+
"loss": 1.4151,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.010282776349614395,
|
| 90 |
+
"grad_norm": 0.1819111704826355,
|
| 91 |
+
"learning_rate": 0.0006,
|
| 92 |
+
"loss": 1.3921,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.011139674378748929,
|
| 97 |
+
"grad_norm": 0.18399009108543396,
|
| 98 |
+
"learning_rate": 0.0006500000000000001,
|
| 99 |
+
"loss": 1.3712,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.011996572407883462,
|
| 104 |
+
"grad_norm": 0.17457200586795807,
|
| 105 |
+
"learning_rate": 0.0007,
|
| 106 |
+
"loss": 1.329,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.012853470437017995,
|
| 111 |
+
"grad_norm": 0.1598544865846634,
|
| 112 |
+
"learning_rate": 0.00075,
|
| 113 |
+
"loss": 1.2791,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.013710368466152529,
|
| 118 |
+
"grad_norm": 0.1446290910243988,
|
| 119 |
+
"learning_rate": 0.0008,
|
| 120 |
+
"loss": 1.1637,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.01456726649528706,
|
| 125 |
+
"grad_norm": 0.130220428109169,
|
| 126 |
+
"learning_rate": 0.00085,
|
| 127 |
+
"loss": 1.1213,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.015424164524421594,
|
| 132 |
+
"grad_norm": 0.13525986671447754,
|
| 133 |
+
"learning_rate": 0.0009000000000000001,
|
| 134 |
+
"loss": 1.0648,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.016281062553556127,
|
| 139 |
+
"grad_norm": 0.13430772721767426,
|
| 140 |
+
"learning_rate": 0.00095,
|
| 141 |
+
"loss": 1.0397,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.01713796058269066,
|
| 146 |
+
"grad_norm": 0.12910906970500946,
|
| 147 |
+
"learning_rate": 0.001,
|
| 148 |
+
"loss": 0.9857,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.017994858611825194,
|
| 153 |
+
"grad_norm": 0.13353696465492249,
|
| 154 |
+
"learning_rate": 0.0009999892908320648,
|
| 155 |
+
"loss": 0.9221,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.018851756640959727,
|
| 160 |
+
"grad_norm": 0.1093161553144455,
|
| 161 |
+
"learning_rate": 0.0009999571637870036,
|
| 162 |
+
"loss": 0.8759,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.01970865467009426,
|
| 167 |
+
"grad_norm": 0.11288490891456604,
|
| 168 |
+
"learning_rate": 0.0009999036202410325,
|
| 169 |
+
"loss": 0.804,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.02056555269922879,
|
| 174 |
+
"grad_norm": 0.104631707072258,
|
| 175 |
+
"learning_rate": 0.0009998286624877785,
|
| 176 |
+
"loss": 0.7662,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.021422450728363324,
|
| 181 |
+
"grad_norm": 0.0997966080904007,
|
| 182 |
+
"learning_rate": 0.0009997322937381828,
|
| 183 |
+
"loss": 0.7223,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.022279348757497857,
|
| 188 |
+
"grad_norm": 0.09641406685113907,
|
| 189 |
+
"learning_rate": 0.0009996145181203615,
|
| 190 |
+
"loss": 0.6928,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.02313624678663239,
|
| 195 |
+
"grad_norm": 0.08281794935464859,
|
| 196 |
+
"learning_rate": 0.00099947534067943,
|
| 197 |
+
"loss": 0.6405,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.023993144815766924,
|
| 202 |
+
"grad_norm": 0.08452638983726501,
|
| 203 |
+
"learning_rate": 0.0009993147673772868,
|
| 204 |
+
"loss": 0.6556,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.024850042844901457,
|
| 209 |
+
"grad_norm": 0.08429277688264847,
|
| 210 |
+
"learning_rate": 0.000999132805092358,
|
| 211 |
+
"loss": 0.594,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.02570694087403599,
|
| 216 |
+
"grad_norm": 0.0761115625500679,
|
| 217 |
+
"learning_rate": 0.0009989294616193018,
|
| 218 |
+
"loss": 0.5755,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.026563838903170524,
|
| 223 |
+
"grad_norm": 0.07302208989858627,
|
| 224 |
+
"learning_rate": 0.000998704745668676,
|
| 225 |
+
"loss": 0.5689,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.027420736932305057,
|
| 230 |
+
"grad_norm": 0.07030151039361954,
|
| 231 |
+
"learning_rate": 0.000998458666866564,
|
| 232 |
+
"loss": 0.5309,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.028277634961439587,
|
| 237 |
+
"grad_norm": 0.06726360321044922,
|
| 238 |
+
"learning_rate": 0.0009981912357541628,
|
| 239 |
+
"loss": 0.522,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.02913453299057412,
|
| 244 |
+
"grad_norm": 0.06689871847629547,
|
| 245 |
+
"learning_rate": 0.0009979024637873308,
|
| 246 |
+
"loss": 0.5076,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.029991431019708654,
|
| 251 |
+
"grad_norm": 0.06583524495363235,
|
| 252 |
+
"learning_rate": 0.0009975923633360985,
|
| 253 |
+
"loss": 0.4853,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.030848329048843187,
|
| 258 |
+
"grad_norm": 0.056236833333969116,
|
| 259 |
+
"learning_rate": 0.0009972609476841367,
|
| 260 |
+
"loss": 0.4725,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.031705227077977724,
|
| 265 |
+
"grad_norm": 0.060708239674568176,
|
| 266 |
+
"learning_rate": 0.0009969082310281891,
|
| 267 |
+
"loss": 0.4617,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.032562125107112254,
|
| 272 |
+
"grad_norm": 0.05682970583438873,
|
| 273 |
+
"learning_rate": 0.0009965342284774632,
|
| 274 |
+
"loss": 0.4696,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.033419023136246784,
|
| 279 |
+
"grad_norm": 0.05259707570075989,
|
| 280 |
+
"learning_rate": 0.0009961389560529835,
|
| 281 |
+
"loss": 0.4548,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.03427592116538132,
|
| 286 |
+
"grad_norm": 0.053453583270311356,
|
| 287 |
+
"learning_rate": 0.0009957224306869053,
|
| 288 |
+
"loss": 0.4514,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.03513281919451585,
|
| 293 |
+
"grad_norm": 0.05121961236000061,
|
| 294 |
+
"learning_rate": 0.0009952846702217886,
|
| 295 |
+
"loss": 0.4327,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.03598971722365039,
|
| 300 |
+
"grad_norm": 0.04787424951791763,
|
| 301 |
+
"learning_rate": 0.0009948256934098352,
|
| 302 |
+
"loss": 0.4184,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.03684661525278492,
|
| 307 |
+
"grad_norm": 0.052464596927165985,
|
| 308 |
+
"learning_rate": 0.0009943455199120836,
|
| 309 |
+
"loss": 0.4039,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.037703513281919454,
|
| 314 |
+
"grad_norm": 0.04593312367796898,
|
| 315 |
+
"learning_rate": 0.0009938441702975688,
|
| 316 |
+
"loss": 0.4027,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.038560411311053984,
|
| 321 |
+
"grad_norm": 0.04613376036286354,
|
| 322 |
+
"learning_rate": 0.0009933216660424397,
|
| 323 |
+
"loss": 0.3833,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.03941730934018852,
|
| 328 |
+
"grad_norm": 0.0438942089676857,
|
| 329 |
+
"learning_rate": 0.0009927780295290389,
|
| 330 |
+
"loss": 0.3943,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.04027420736932305,
|
| 335 |
+
"grad_norm": 0.03944185748696327,
|
| 336 |
+
"learning_rate": 0.0009922132840449458,
|
| 337 |
+
"loss": 0.3751,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.04113110539845758,
|
| 342 |
+
"grad_norm": 0.0458449125289917,
|
| 343 |
+
"learning_rate": 0.0009916274537819774,
|
| 344 |
+
"loss": 0.3691,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.04198800342759212,
|
| 349 |
+
"grad_norm": 0.040158357471227646,
|
| 350 |
+
"learning_rate": 0.000991020563835152,
|
| 351 |
+
"loss": 0.3793,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.04284490145672665,
|
| 356 |
+
"grad_norm": 0.044870492070913315,
|
| 357 |
+
"learning_rate": 0.0009903926402016153,
|
| 358 |
+
"loss": 0.3533,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.043701799485861184,
|
| 363 |
+
"grad_norm": 0.037956662476062775,
|
| 364 |
+
"learning_rate": 0.0009897437097795257,
|
| 365 |
+
"loss": 0.3773,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.044558697514995714,
|
| 370 |
+
"grad_norm": 0.03750976547598839,
|
| 371 |
+
"learning_rate": 0.0009890738003669028,
|
| 372 |
+
"loss": 0.3547,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.04541559554413025,
|
| 377 |
+
"grad_norm": 0.03590930253267288,
|
| 378 |
+
"learning_rate": 0.0009883829406604362,
|
| 379 |
+
"loss": 0.3447,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.04627249357326478,
|
| 384 |
+
"grad_norm": 0.03637269139289856,
|
| 385 |
+
"learning_rate": 0.0009876711602542563,
|
| 386 |
+
"loss": 0.3557,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.04712939160239932,
|
| 391 |
+
"grad_norm": 0.0436246357858181,
|
| 392 |
+
"learning_rate": 0.0009869384896386668,
|
| 393 |
+
"loss": 0.3361,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.04798628963153385,
|
| 398 |
+
"grad_norm": 0.03389494866132736,
|
| 399 |
+
"learning_rate": 0.0009861849601988384,
|
| 400 |
+
"loss": 0.3229,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.04884318766066838,
|
| 405 |
+
"grad_norm": 0.03282163292169571,
|
| 406 |
+
"learning_rate": 0.0009854106042134641,
|
| 407 |
+
"loss": 0.2967,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.049700085689802914,
|
| 412 |
+
"grad_norm": 0.03293761610984802,
|
| 413 |
+
"learning_rate": 0.0009846154548533773,
|
| 414 |
+
"loss": 0.3236,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.050556983718937444,
|
| 419 |
+
"grad_norm": 0.03896806761622429,
|
| 420 |
+
"learning_rate": 0.0009837995461801298,
|
| 421 |
+
"loss": 0.3117,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.05141388174807198,
|
| 426 |
+
"grad_norm": 0.03278586268424988,
|
| 427 |
+
"learning_rate": 0.0009829629131445341,
|
| 428 |
+
"loss": 0.3068,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.05227077977720651,
|
| 433 |
+
"grad_norm": 0.0321279913187027,
|
| 434 |
+
"learning_rate": 0.0009821055915851646,
|
| 435 |
+
"loss": 0.3118,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.05312767780634105,
|
| 440 |
+
"grad_norm": 0.03264940157532692,
|
| 441 |
+
"learning_rate": 0.0009812276182268236,
|
| 442 |
+
"loss": 0.3125,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.05398457583547558,
|
| 447 |
+
"grad_norm": 0.03603900223970413,
|
| 448 |
+
"learning_rate": 0.0009803290306789677,
|
| 449 |
+
"loss": 0.3,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.054841473864610114,
|
| 454 |
+
"grad_norm": 0.029965505003929138,
|
| 455 |
+
"learning_rate": 0.0009794098674340967,
|
| 456 |
+
"loss": 0.3096,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.055698371893744644,
|
| 461 |
+
"grad_norm": 0.036456797271966934,
|
| 462 |
+
"learning_rate": 0.0009784701678661044,
|
| 463 |
+
"loss": 0.2901,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.056555269922879174,
|
| 468 |
+
"grad_norm": 0.032042618840932846,
|
| 469 |
+
"learning_rate": 0.0009775099722285933,
|
| 470 |
+
"loss": 0.2864,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.05741216795201371,
|
| 475 |
+
"grad_norm": 0.03345092013478279,
|
| 476 |
+
"learning_rate": 0.0009765293216531485,
|
| 477 |
+
"loss": 0.2854,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.05826906598114824,
|
| 482 |
+
"grad_norm": 0.029210660606622696,
|
| 483 |
+
"learning_rate": 0.0009755282581475768,
|
| 484 |
+
"loss": 0.2862,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.05912596401028278,
|
| 489 |
+
"grad_norm": 0.02906818874180317,
|
| 490 |
+
"learning_rate": 0.000974506824594107,
|
| 491 |
+
"loss": 0.2756,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.05998286203941731,
|
| 496 |
+
"grad_norm": 0.032076746225357056,
|
| 497 |
+
"learning_rate": 0.0009734650647475529,
|
| 498 |
+
"loss": 0.274,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.060839760068551844,
|
| 503 |
+
"grad_norm": 0.02805478870868683,
|
| 504 |
+
"learning_rate": 0.0009724030232334391,
|
| 505 |
+
"loss": 0.2807,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.061696658097686374,
|
| 510 |
+
"grad_norm": 0.027483096346259117,
|
| 511 |
+
"learning_rate": 0.0009713207455460893,
|
| 512 |
+
"loss": 0.2657,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.06255355612682091,
|
| 517 |
+
"grad_norm": 0.026773786172270775,
|
| 518 |
+
"learning_rate": 0.0009702182780466775,
|
| 519 |
+
"loss": 0.2638,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.06341045415595545,
|
| 524 |
+
"grad_norm": 0.025541041046380997,
|
| 525 |
+
"learning_rate": 0.0009690956679612422,
|
| 526 |
+
"loss": 0.2621,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.06426735218508997,
|
| 531 |
+
"grad_norm": 0.028647374361753464,
|
| 532 |
+
"learning_rate": 0.0009679529633786629,
|
| 533 |
+
"loss": 0.268,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.06512425021422451,
|
| 538 |
+
"grad_norm": 0.025853468105196953,
|
| 539 |
+
"learning_rate": 0.0009667902132486009,
|
| 540 |
+
"loss": 0.2665,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.06598114824335904,
|
| 545 |
+
"grad_norm": 0.02473859116435051,
|
| 546 |
+
"learning_rate": 0.0009656074673794017,
|
| 547 |
+
"loss": 0.2633,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.06683804627249357,
|
| 552 |
+
"grad_norm": 0.027110164985060692,
|
| 553 |
+
"learning_rate": 0.0009644047764359622,
|
| 554 |
+
"loss": 0.2591,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.0676949443016281,
|
| 559 |
+
"grad_norm": 0.02725694142282009,
|
| 560 |
+
"learning_rate": 0.0009631821919375591,
|
| 561 |
+
"loss": 0.2486,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.06855184233076264,
|
| 566 |
+
"grad_norm": 0.023004408925771713,
|
| 567 |
+
"learning_rate": 0.0009619397662556434,
|
| 568 |
+
"loss": 0.2477,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.06940874035989718,
|
| 573 |
+
"grad_norm": 0.03091372549533844,
|
| 574 |
+
"learning_rate": 0.0009606775526115963,
|
| 575 |
+
"loss": 0.2625,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.0702656383890317,
|
| 580 |
+
"grad_norm": 0.02504062093794346,
|
| 581 |
+
"learning_rate": 0.0009593956050744492,
|
| 582 |
+
"loss": 0.2603,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.07112253641816624,
|
| 587 |
+
"grad_norm": 0.028378132730722427,
|
| 588 |
+
"learning_rate": 0.0009580939785585681,
|
| 589 |
+
"loss": 0.2501,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.07197943444730077,
|
| 594 |
+
"grad_norm": 0.031088434159755707,
|
| 595 |
+
"learning_rate": 0.0009567727288213005,
|
| 596 |
+
"loss": 0.2463,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.0728363324764353,
|
| 601 |
+
"grad_norm": 0.02509693056344986,
|
| 602 |
+
"learning_rate": 0.000955431912460588,
|
| 603 |
+
"loss": 0.2619,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.07369323050556983,
|
| 608 |
+
"grad_norm": 0.024940533563494682,
|
| 609 |
+
"learning_rate": 0.0009540715869125407,
|
| 610 |
+
"loss": 0.2565,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.07455012853470437,
|
| 615 |
+
"grad_norm": 0.026020990684628487,
|
| 616 |
+
"learning_rate": 0.0009526918104489777,
|
| 617 |
+
"loss": 0.2436,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.07540702656383891,
|
| 622 |
+
"grad_norm": 0.023458324372768402,
|
| 623 |
+
"learning_rate": 0.0009512926421749304,
|
| 624 |
+
"loss": 0.2362,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.07626392459297343,
|
| 629 |
+
"grad_norm": 0.0243577491492033,
|
| 630 |
+
"learning_rate": 0.0009498741420261108,
|
| 631 |
+
"loss": 0.243,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.07712082262210797,
|
| 636 |
+
"grad_norm": 0.0244216900318861,
|
| 637 |
+
"learning_rate": 0.0009484363707663442,
|
| 638 |
+
"loss": 0.2378,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.0779777206512425,
|
| 643 |
+
"grad_norm": 0.02734869159758091,
|
| 644 |
+
"learning_rate": 0.0009469793899849661,
|
| 645 |
+
"loss": 0.2344,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.07883461868037704,
|
| 650 |
+
"grad_norm": 0.022375451400876045,
|
| 651 |
+
"learning_rate": 0.0009455032620941839,
|
| 652 |
+
"loss": 0.239,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.07969151670951156,
|
| 657 |
+
"grad_norm": 0.026033056899905205,
|
| 658 |
+
"learning_rate": 0.0009440080503264037,
|
| 659 |
+
"loss": 0.2334,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.0805484147386461,
|
| 664 |
+
"grad_norm": 0.022317685186862946,
|
| 665 |
+
"learning_rate": 0.0009424938187315209,
|
| 666 |
+
"loss": 0.2327,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.08140531276778064,
|
| 671 |
+
"grad_norm": 0.02240647003054619,
|
| 672 |
+
"learning_rate": 0.0009409606321741775,
|
| 673 |
+
"loss": 0.2335,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.08226221079691516,
|
| 678 |
+
"grad_norm": 0.022393332794308662,
|
| 679 |
+
"learning_rate": 0.0009394085563309827,
|
| 680 |
+
"loss": 0.2237,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.0831191088260497,
|
| 685 |
+
"grad_norm": 0.02859143353998661,
|
| 686 |
+
"learning_rate": 0.0009378376576876999,
|
| 687 |
+
"loss": 0.2243,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.08397600685518423,
|
| 692 |
+
"grad_norm": 0.02249022200703621,
|
| 693 |
+
"learning_rate": 0.0009362480035363986,
|
| 694 |
+
"loss": 0.2348,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.08483290488431877,
|
| 699 |
+
"grad_norm": 0.031574804335832596,
|
| 700 |
+
"learning_rate": 0.0009346396619725719,
|
| 701 |
+
"loss": 0.2209,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.0856898029134533,
|
| 706 |
+
"grad_norm": 0.024664921686053276,
|
| 707 |
+
"learning_rate": 0.0009330127018922195,
|
| 708 |
+
"loss": 0.2437,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.08654670094258783,
|
| 713 |
+
"grad_norm": 0.023365097120404243,
|
| 714 |
+
"learning_rate": 0.0009313671929888959,
|
| 715 |
+
"loss": 0.2271,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.08740359897172237,
|
| 720 |
+
"grad_norm": 0.02179112657904625,
|
| 721 |
+
"learning_rate": 0.0009297032057507264,
|
| 722 |
+
"loss": 0.2251,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.08826049700085689,
|
| 727 |
+
"grad_norm": 0.02501068077981472,
|
| 728 |
+
"learning_rate": 0.0009280208114573858,
|
| 729 |
+
"loss": 0.2177,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.08911739502999143,
|
| 734 |
+
"grad_norm": 0.022900037467479706,
|
| 735 |
+
"learning_rate": 0.0009263200821770461,
|
| 736 |
+
"loss": 0.2317,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.08997429305912596,
|
| 741 |
+
"grad_norm": 0.02185271680355072,
|
| 742 |
+
"learning_rate": 0.0009246010907632895,
|
| 743 |
+
"loss": 0.2209,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.0908311910882605,
|
| 748 |
+
"grad_norm": 0.021373869851231575,
|
| 749 |
+
"learning_rate": 0.0009228639108519867,
|
| 750 |
+
"loss": 0.2261,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.09168808911739502,
|
| 755 |
+
"grad_norm": 0.02603291906416416,
|
| 756 |
+
"learning_rate": 0.0009211086168581433,
|
| 757 |
+
"loss": 0.2212,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.09254498714652956,
|
| 762 |
+
"grad_norm": 0.021860316395759583,
|
| 763 |
+
"learning_rate": 0.0009193352839727121,
|
| 764 |
+
"loss": 0.2163,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.0934018851756641,
|
| 769 |
+
"grad_norm": 0.021575644612312317,
|
| 770 |
+
"learning_rate": 0.0009175439881593715,
|
| 771 |
+
"loss": 0.2252,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.09425878320479864,
|
| 776 |
+
"grad_norm": 0.02140972390770912,
|
| 777 |
+
"learning_rate": 0.0009157348061512727,
|
| 778 |
+
"loss": 0.2076,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.09511568123393316,
|
| 783 |
+
"grad_norm": 0.023880021646618843,
|
| 784 |
+
"learning_rate": 0.0009139078154477511,
|
| 785 |
+
"loss": 0.2147,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.0959725792630677,
|
| 790 |
+
"grad_norm": 0.02249889075756073,
|
| 791 |
+
"learning_rate": 0.0009120630943110077,
|
| 792 |
+
"loss": 0.2047,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.09682947729220223,
|
| 797 |
+
"grad_norm": 0.020879626274108887,
|
| 798 |
+
"learning_rate": 0.0009102007217627568,
|
| 799 |
+
"loss": 0.2231,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.09768637532133675,
|
| 804 |
+
"grad_norm": 0.021978724747896194,
|
| 805 |
+
"learning_rate": 0.0009083207775808396,
|
| 806 |
+
"loss": 0.2145,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.09854327335047129,
|
| 811 |
+
"grad_norm": 0.023305930197238922,
|
| 812 |
+
"learning_rate": 0.0009064233422958076,
|
| 813 |
+
"loss": 0.2156,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.09940017137960583,
|
| 818 |
+
"grad_norm": 0.022691812366247177,
|
| 819 |
+
"learning_rate": 0.0009045084971874737,
|
| 820 |
+
"loss": 0.2142,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.10025706940874037,
|
| 825 |
+
"grad_norm": 0.019997967407107353,
|
| 826 |
+
"learning_rate": 0.0009025763242814291,
|
| 827 |
+
"loss": 0.2141,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.10111396743787489,
|
| 832 |
+
"grad_norm": 0.021491916850209236,
|
| 833 |
+
"learning_rate": 0.0009006269063455304,
|
| 834 |
+
"loss": 0.2095,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.10197086546700942,
|
| 839 |
+
"grad_norm": 0.019741587340831757,
|
| 840 |
+
"learning_rate": 0.0008986603268863536,
|
| 841 |
+
"loss": 0.2213,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.10282776349614396,
|
| 846 |
+
"grad_norm": 0.02245885692536831,
|
| 847 |
+
"learning_rate": 0.0008966766701456176,
|
| 848 |
+
"loss": 0.2041,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.1036846615252785,
|
| 853 |
+
"grad_norm": 0.020367003977298737,
|
| 854 |
+
"learning_rate": 0.000894676021096575,
|
| 855 |
+
"loss": 0.2052,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.10454155955441302,
|
| 860 |
+
"grad_norm": 0.021478816866874695,
|
| 861 |
+
"learning_rate": 0.0008926584654403724,
|
| 862 |
+
"loss": 0.1999,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.10539845758354756,
|
| 867 |
+
"grad_norm": 0.022485392168164253,
|
| 868 |
+
"learning_rate": 0.0008906240896023794,
|
| 869 |
+
"loss": 0.2066,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.1062553556126821,
|
| 874 |
+
"grad_norm": 0.02136247418820858,
|
| 875 |
+
"learning_rate": 0.0008885729807284854,
|
| 876 |
+
"loss": 0.1972,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.10711225364181662,
|
| 881 |
+
"grad_norm": 0.020799005404114723,
|
| 882 |
+
"learning_rate": 0.0008865052266813684,
|
| 883 |
+
"loss": 0.2062,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.10796915167095116,
|
| 888 |
+
"grad_norm": 0.02335492894053459,
|
| 889 |
+
"learning_rate": 0.0008844209160367298,
|
| 890 |
+
"loss": 0.2004,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.10882604970008569,
|
| 895 |
+
"grad_norm": 0.025448890402913094,
|
| 896 |
+
"learning_rate": 0.0008823201380795002,
|
| 897 |
+
"loss": 0.2053,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.10968294772922023,
|
| 902 |
+
"grad_norm": 0.024005141109228134,
|
| 903 |
+
"learning_rate": 0.0008802029828000156,
|
| 904 |
+
"loss": 0.1926,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.11053984575835475,
|
| 909 |
+
"grad_norm": 0.021053491160273552,
|
| 910 |
+
"learning_rate": 0.0008780695408901613,
|
| 911 |
+
"loss": 0.1938,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.11139674378748929,
|
| 916 |
+
"grad_norm": 0.023733297362923622,
|
| 917 |
+
"learning_rate": 0.0008759199037394887,
|
| 918 |
+
"loss": 0.2013,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.11225364181662383,
|
| 923 |
+
"grad_norm": 0.021928993985056877,
|
| 924 |
+
"learning_rate": 0.0008737541634312985,
|
| 925 |
+
"loss": 0.1939,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.11311053984575835,
|
| 930 |
+
"grad_norm": 0.02457588165998459,
|
| 931 |
+
"learning_rate": 0.0008715724127386971,
|
| 932 |
+
"loss": 0.2005,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.11396743787489289,
|
| 937 |
+
"grad_norm": 0.021438535302877426,
|
| 938 |
+
"learning_rate": 0.0008693747451206231,
|
| 939 |
+
"loss": 0.1984,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.11482433590402742,
|
| 944 |
+
"grad_norm": 0.025149062275886536,
|
| 945 |
+
"learning_rate": 0.0008671612547178428,
|
| 946 |
+
"loss": 0.2042,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.11568123393316196,
|
| 951 |
+
"grad_norm": 0.029300540685653687,
|
| 952 |
+
"learning_rate": 0.0008649320363489178,
|
| 953 |
+
"loss": 0.2056,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.11653813196229648,
|
| 958 |
+
"grad_norm": 0.022236980497837067,
|
| 959 |
+
"learning_rate": 0.0008626871855061438,
|
| 960 |
+
"loss": 0.2056,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.11739502999143102,
|
| 965 |
+
"grad_norm": 0.02103867195546627,
|
| 966 |
+
"learning_rate": 0.0008604267983514594,
|
| 967 |
+
"loss": 0.1949,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.11825192802056556,
|
| 972 |
+
"grad_norm": 0.025057192891836166,
|
| 973 |
+
"learning_rate": 0.0008581509717123273,
|
| 974 |
+
"loss": 0.1993,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.11910882604970009,
|
| 979 |
+
"grad_norm": 0.021641414612531662,
|
| 980 |
+
"learning_rate": 0.0008558598030775857,
|
| 981 |
+
"loss": 0.1969,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.11996572407883462,
|
| 986 |
+
"grad_norm": 0.022903475910425186,
|
| 987 |
+
"learning_rate": 0.0008535533905932737,
|
| 988 |
+
"loss": 0.1971,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.12082262210796915,
|
| 993 |
+
"grad_norm": 0.021558206528425217,
|
| 994 |
+
"learning_rate": 0.0008512318330584259,
|
| 995 |
+
"loss": 0.1932,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.12167952013710369,
|
| 1000 |
+
"grad_norm": 0.024407191202044487,
|
| 1001 |
+
"learning_rate": 0.0008488952299208401,
|
| 1002 |
+
"loss": 0.1872,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.12253641816623821,
|
| 1007 |
+
"grad_norm": 0.02087639644742012,
|
| 1008 |
+
"learning_rate": 0.000846543681272818,
|
| 1009 |
+
"loss": 0.2032,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.12339331619537275,
|
| 1014 |
+
"grad_norm": 0.02548481896519661,
|
| 1015 |
+
"learning_rate": 0.000844177287846877,
|
| 1016 |
+
"loss": 0.1924,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.12425021422450729,
|
| 1021 |
+
"grad_norm": 0.02311697043478489,
|
| 1022 |
+
"learning_rate": 0.0008417961510114356,
|
| 1023 |
+
"loss": 0.1925,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.12510711225364182,
|
| 1028 |
+
"grad_norm": 0.018741579726338387,
|
| 1029 |
+
"learning_rate": 0.0008394003727664709,
|
| 1030 |
+
"loss": 0.1812,
|
| 1031 |
+
"step": 146
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.12596401028277635,
|
| 1035 |
+
"grad_norm": 0.025283221155405045,
|
| 1036 |
+
"learning_rate": 0.000836990055739149,
|
| 1037 |
+
"loss": 0.1888,
|
| 1038 |
+
"step": 147
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.1268209083119109,
|
| 1042 |
+
"grad_norm": 0.025587448850274086,
|
| 1043 |
+
"learning_rate": 0.0008345653031794292,
|
| 1044 |
+
"loss": 0.1942,
|
| 1045 |
+
"step": 148
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.12767780634104542,
|
| 1049 |
+
"grad_norm": 0.020198052749037743,
|
| 1050 |
+
"learning_rate": 0.0008321262189556409,
|
| 1051 |
+
"loss": 0.1863,
|
| 1052 |
+
"step": 149
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.12853470437017994,
|
| 1056 |
+
"grad_norm": 0.02162170223891735,
|
| 1057 |
+
"learning_rate": 0.0008296729075500344,
|
| 1058 |
+
"loss": 0.1891,
|
| 1059 |
+
"step": 150
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.1293916023993145,
|
| 1063 |
+
"grad_norm": 0.022257110103964806,
|
| 1064 |
+
"learning_rate": 0.0008272054740543053,
|
| 1065 |
+
"loss": 0.194,
|
| 1066 |
+
"step": 151
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.13024850042844902,
|
| 1070 |
+
"grad_norm": 0.022947294637560844,
|
| 1071 |
+
"learning_rate": 0.0008247240241650918,
|
| 1072 |
+
"loss": 0.1836,
|
| 1073 |
+
"step": 152
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.13110539845758354,
|
| 1077 |
+
"grad_norm": 0.021060343831777573,
|
| 1078 |
+
"learning_rate": 0.0008222286641794488,
|
| 1079 |
+
"loss": 0.1886,
|
| 1080 |
+
"step": 153
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.1319622964867181,
|
| 1084 |
+
"grad_norm": 0.025356875732541084,
|
| 1085 |
+
"learning_rate": 0.0008197195009902923,
|
| 1086 |
+
"loss": 0.1875,
|
| 1087 |
+
"step": 154
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.1328191945158526,
|
| 1091 |
+
"grad_norm": 0.022868521511554718,
|
| 1092 |
+
"learning_rate": 0.0008171966420818228,
|
| 1093 |
+
"loss": 0.1904,
|
| 1094 |
+
"step": 155
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.13367609254498714,
|
| 1098 |
+
"grad_norm": 0.023869860917329788,
|
| 1099 |
+
"learning_rate": 0.0008146601955249188,
|
| 1100 |
+
"loss": 0.1832,
|
| 1101 |
+
"step": 156
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.13453299057412169,
|
| 1105 |
+
"grad_norm": 0.022344481199979782,
|
| 1106 |
+
"learning_rate": 0.0008121102699725089,
|
| 1107 |
+
"loss": 0.1814,
|
| 1108 |
+
"step": 157
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.1353898886032562,
|
| 1112 |
+
"grad_norm": 0.0211899783462286,
|
| 1113 |
+
"learning_rate": 0.0008095469746549171,
|
| 1114 |
+
"loss": 0.1929,
|
| 1115 |
+
"step": 158
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.13624678663239073,
|
| 1119 |
+
"grad_norm": 0.022394029423594475,
|
| 1120 |
+
"learning_rate": 0.0008069704193751832,
|
| 1121 |
+
"loss": 0.1879,
|
| 1122 |
+
"step": 159
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.13710368466152528,
|
| 1126 |
+
"grad_norm": 0.019436044618487358,
|
| 1127 |
+
"learning_rate": 0.0008043807145043603,
|
| 1128 |
+
"loss": 0.187,
|
| 1129 |
+
"step": 160
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.1379605826906598,
|
| 1133 |
+
"grad_norm": 0.028875339776277542,
|
| 1134 |
+
"learning_rate": 0.0008017779709767858,
|
| 1135 |
+
"loss": 0.1853,
|
| 1136 |
+
"step": 161
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.13881748071979436,
|
| 1140 |
+
"grad_norm": 0.024842459708452225,
|
| 1141 |
+
"learning_rate": 0.0007991623002853296,
|
| 1142 |
+
"loss": 0.1858,
|
| 1143 |
+
"step": 162
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.13967437874892888,
|
| 1147 |
+
"grad_norm": 0.021923067048192024,
|
| 1148 |
+
"learning_rate": 0.0007965338144766185,
|
| 1149 |
+
"loss": 0.1879,
|
| 1150 |
+
"step": 163
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.1405312767780634,
|
| 1154 |
+
"grad_norm": 0.020001094788312912,
|
| 1155 |
+
"learning_rate": 0.0007938926261462366,
|
| 1156 |
+
"loss": 0.1764,
|
| 1157 |
+
"step": 164
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.14138817480719795,
|
| 1161 |
+
"grad_norm": 0.02206343039870262,
|
| 1162 |
+
"learning_rate": 0.0007912388484339011,
|
| 1163 |
+
"loss": 0.1851,
|
| 1164 |
+
"step": 165
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.14224507283633248,
|
| 1168 |
+
"grad_norm": 0.018592318519949913,
|
| 1169 |
+
"learning_rate": 0.0007885725950186169,
|
| 1170 |
+
"loss": 0.1807,
|
| 1171 |
+
"step": 166
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.143101970865467,
|
| 1175 |
+
"grad_norm": 0.022569900378584862,
|
| 1176 |
+
"learning_rate": 0.000785893980113806,
|
| 1177 |
+
"loss": 0.1782,
|
| 1178 |
+
"step": 167
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.14395886889460155,
|
| 1182 |
+
"grad_norm": 0.02649298682808876,
|
| 1183 |
+
"learning_rate": 0.0007832031184624164,
|
| 1184 |
+
"loss": 0.1796,
|
| 1185 |
+
"step": 168
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.14481576692373607,
|
| 1189 |
+
"grad_norm": 0.018016191199421883,
|
| 1190 |
+
"learning_rate": 0.000780500125332005,
|
| 1191 |
+
"loss": 0.1773,
|
| 1192 |
+
"step": 169
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.1456726649528706,
|
| 1196 |
+
"grad_norm": 0.02014097571372986,
|
| 1197 |
+
"learning_rate": 0.0007777851165098011,
|
| 1198 |
+
"loss": 0.1746,
|
| 1199 |
+
"step": 170
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.14652956298200515,
|
| 1203 |
+
"grad_norm": 0.01795336976647377,
|
| 1204 |
+
"learning_rate": 0.0007750582082977468,
|
| 1205 |
+
"loss": 0.1794,
|
| 1206 |
+
"step": 171
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.14738646101113967,
|
| 1210 |
+
"grad_norm": 0.02241736464202404,
|
| 1211 |
+
"learning_rate": 0.0007723195175075137,
|
| 1212 |
+
"loss": 0.1697,
|
| 1213 |
+
"step": 172
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.14824335904027422,
|
| 1217 |
+
"grad_norm": 0.021646304056048393,
|
| 1218 |
+
"learning_rate": 0.0007695691614555002,
|
| 1219 |
+
"loss": 0.1767,
|
| 1220 |
+
"step": 173
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.14910025706940874,
|
| 1224 |
+
"grad_norm": 0.01908080466091633,
|
| 1225 |
+
"learning_rate": 0.0007668072579578058,
|
| 1226 |
+
"loss": 0.1808,
|
| 1227 |
+
"step": 174
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.14995715509854327,
|
| 1231 |
+
"grad_norm": 0.019076567143201828,
|
| 1232 |
+
"learning_rate": 0.000764033925325184,
|
| 1233 |
+
"loss": 0.1753,
|
| 1234 |
+
"step": 175
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.15081405312767782,
|
| 1238 |
+
"grad_norm": 0.01973322220146656,
|
| 1239 |
+
"learning_rate": 0.0007612492823579744,
|
| 1240 |
+
"loss": 0.1747,
|
| 1241 |
+
"step": 176
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.15167095115681234,
|
| 1245 |
+
"grad_norm": 0.017605643719434738,
|
| 1246 |
+
"learning_rate": 0.0007584534483410137,
|
| 1247 |
+
"loss": 0.1746,
|
| 1248 |
+
"step": 177
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.15252784918594686,
|
| 1252 |
+
"grad_norm": 0.017287466675043106,
|
| 1253 |
+
"learning_rate": 0.0007556465430385259,
|
| 1254 |
+
"loss": 0.1778,
|
| 1255 |
+
"step": 178
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.1533847472150814,
|
| 1259 |
+
"grad_norm": 0.019033176824450493,
|
| 1260 |
+
"learning_rate": 0.0007528286866889924,
|
| 1261 |
+
"loss": 0.1795,
|
| 1262 |
+
"step": 179
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.15424164524421594,
|
| 1266 |
+
"grad_norm": 0.019875552505254745,
|
| 1267 |
+
"learning_rate": 0.00075,
|
| 1268 |
+
"loss": 0.1637,
|
| 1269 |
+
"step": 180
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.15509854327335046,
|
| 1273 |
+
"grad_norm": 0.018225200474262238,
|
| 1274 |
+
"learning_rate": 0.0007471606041430723,
|
| 1275 |
+
"loss": 0.167,
|
| 1276 |
+
"step": 181
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.155955441302485,
|
| 1280 |
+
"grad_norm": 0.019243160262703896,
|
| 1281 |
+
"learning_rate": 0.0007443106207484776,
|
| 1282 |
+
"loss": 0.1704,
|
| 1283 |
+
"step": 182
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.15681233933161953,
|
| 1287 |
+
"grad_norm": 0.019462725147604942,
|
| 1288 |
+
"learning_rate": 0.0007414501719000186,
|
| 1289 |
+
"loss": 0.1764,
|
| 1290 |
+
"step": 183
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.15766923736075408,
|
| 1294 |
+
"grad_norm": 0.0178945641964674,
|
| 1295 |
+
"learning_rate": 0.0007385793801298042,
|
| 1296 |
+
"loss": 0.1771,
|
| 1297 |
+
"step": 184
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.1585261353898886,
|
| 1301 |
+
"grad_norm": 0.01750069111585617,
|
| 1302 |
+
"learning_rate": 0.000735698368412999,
|
| 1303 |
+
"loss": 0.1773,
|
| 1304 |
+
"step": 185
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.15938303341902313,
|
| 1308 |
+
"grad_norm": 0.023665515705943108,
|
| 1309 |
+
"learning_rate": 0.0007328072601625557,
|
| 1310 |
+
"loss": 0.1842,
|
| 1311 |
+
"step": 186
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.16023993144815768,
|
| 1315 |
+
"grad_norm": 0.021700644865632057,
|
| 1316 |
+
"learning_rate": 0.00072990617922393,
|
| 1317 |
+
"loss": 0.1668,
|
| 1318 |
+
"step": 187
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.1610968294772922,
|
| 1322 |
+
"grad_norm": 0.022405751049518585,
|
| 1323 |
+
"learning_rate": 0.0007269952498697733,
|
| 1324 |
+
"loss": 0.1831,
|
| 1325 |
+
"step": 188
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.16195372750642673,
|
| 1329 |
+
"grad_norm": 0.02437387965619564,
|
| 1330 |
+
"learning_rate": 0.0007240745967946113,
|
| 1331 |
+
"loss": 0.1805,
|
| 1332 |
+
"step": 189
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.16281062553556128,
|
| 1336 |
+
"grad_norm": 0.019709205254912376,
|
| 1337 |
+
"learning_rate": 0.0007211443451095007,
|
| 1338 |
+
"loss": 0.1691,
|
| 1339 |
+
"step": 190
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.1636675235646958,
|
| 1343 |
+
"grad_norm": 0.02336045354604721,
|
| 1344 |
+
"learning_rate": 0.000718204620336671,
|
| 1345 |
+
"loss": 0.1751,
|
| 1346 |
+
"step": 191
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.16452442159383032,
|
| 1350 |
+
"grad_norm": 0.0230086762458086,
|
| 1351 |
+
"learning_rate": 0.0007152555484041476,
|
| 1352 |
+
"loss": 0.1697,
|
| 1353 |
+
"step": 192
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.16538131962296487,
|
| 1357 |
+
"grad_norm": 0.01794915646314621,
|
| 1358 |
+
"learning_rate": 0.0007122972556403566,
|
| 1359 |
+
"loss": 0.1704,
|
| 1360 |
+
"step": 193
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.1662382176520994,
|
| 1364 |
+
"grad_norm": 0.02200886234641075,
|
| 1365 |
+
"learning_rate": 0.0007093298687687141,
|
| 1366 |
+
"loss": 0.1741,
|
| 1367 |
+
"step": 194
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.16709511568123395,
|
| 1371 |
+
"grad_norm": 0.02043880894780159,
|
| 1372 |
+
"learning_rate": 0.0007063535149021973,
|
| 1373 |
+
"loss": 0.1753,
|
| 1374 |
+
"step": 195
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.16795201371036847,
|
| 1378 |
+
"grad_norm": 0.018617108464241028,
|
| 1379 |
+
"learning_rate": 0.0007033683215379002,
|
| 1380 |
+
"loss": 0.1727,
|
| 1381 |
+
"step": 196
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.168808911739503,
|
| 1385 |
+
"grad_norm": 0.023497916758060455,
|
| 1386 |
+
"learning_rate": 0.0007003744165515704,
|
| 1387 |
+
"loss": 0.1647,
|
| 1388 |
+
"step": 197
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.16966580976863754,
|
| 1392 |
+
"grad_norm": 0.021086974069476128,
|
| 1393 |
+
"learning_rate": 0.0006973719281921336,
|
| 1394 |
+
"loss": 0.1675,
|
| 1395 |
+
"step": 198
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.17052270779777207,
|
| 1399 |
+
"grad_norm": 0.018384108319878578,
|
| 1400 |
+
"learning_rate": 0.0006943609850761978,
|
| 1401 |
+
"loss": 0.1738,
|
| 1402 |
+
"step": 199
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.1713796058269066,
|
| 1406 |
+
"grad_norm": 0.019381960853934288,
|
| 1407 |
+
"learning_rate": 0.000691341716182545,
|
| 1408 |
+
"loss": 0.1711,
|
| 1409 |
+
"step": 200
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.17223650385604114,
|
| 1413 |
+
"grad_norm": 0.023446090519428253,
|
| 1414 |
+
"learning_rate": 0.0006883142508466054,
|
| 1415 |
+
"loss": 0.17,
|
| 1416 |
+
"step": 201
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.17309340188517566,
|
| 1420 |
+
"grad_norm": 0.018001163378357887,
|
| 1421 |
+
"learning_rate": 0.0006852787187549182,
|
| 1422 |
+
"loss": 0.1675,
|
| 1423 |
+
"step": 202
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.17395029991431019,
|
| 1427 |
+
"grad_norm": 0.018448330461978912,
|
| 1428 |
+
"learning_rate": 0.000682235249939575,
|
| 1429 |
+
"loss": 0.1732,
|
| 1430 |
+
"step": 203
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.17480719794344474,
|
| 1434 |
+
"grad_norm": 0.022160930559039116,
|
| 1435 |
+
"learning_rate": 0.0006791839747726501,
|
| 1436 |
+
"loss": 0.1737,
|
| 1437 |
+
"step": 204
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.17566409597257926,
|
| 1441 |
+
"grad_norm": 0.021703101694583893,
|
| 1442 |
+
"learning_rate": 0.0006761250239606168,
|
| 1443 |
+
"loss": 0.1716,
|
| 1444 |
+
"step": 205
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.17652099400171378,
|
| 1448 |
+
"grad_norm": 0.03577428311109543,
|
| 1449 |
+
"learning_rate": 0.0006730585285387465,
|
| 1450 |
+
"loss": 0.2503,
|
| 1451 |
+
"step": 206
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.17737789203084833,
|
| 1455 |
+
"grad_norm": 0.021015428006649017,
|
| 1456 |
+
"learning_rate": 0.000669984619865497,
|
| 1457 |
+
"loss": 0.1683,
|
| 1458 |
+
"step": 207
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.17823479005998286,
|
| 1462 |
+
"grad_norm": 0.027599437162280083,
|
| 1463 |
+
"learning_rate": 0.0006669034296168854,
|
| 1464 |
+
"loss": 0.1679,
|
| 1465 |
+
"step": 208
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.1790916880891174,
|
| 1469 |
+
"grad_norm": 0.02527066133916378,
|
| 1470 |
+
"learning_rate": 0.0006638150897808468,
|
| 1471 |
+
"loss": 0.1711,
|
| 1472 |
+
"step": 209
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.17994858611825193,
|
| 1476 |
+
"grad_norm": 0.021573588252067566,
|
| 1477 |
+
"learning_rate": 0.0006607197326515808,
|
| 1478 |
+
"loss": 0.1709,
|
| 1479 |
+
"step": 210
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.18080548414738645,
|
| 1483 |
+
"grad_norm": 0.022143971174955368,
|
| 1484 |
+
"learning_rate": 0.0006576174908238849,
|
| 1485 |
+
"loss": 0.1695,
|
| 1486 |
+
"step": 211
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.181662382176521,
|
| 1490 |
+
"grad_norm": 0.019489064812660217,
|
| 1491 |
+
"learning_rate": 0.0006545084971874737,
|
| 1492 |
+
"loss": 0.1713,
|
| 1493 |
+
"step": 212
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.18251928020565553,
|
| 1497 |
+
"grad_norm": 0.016801459714770317,
|
| 1498 |
+
"learning_rate": 0.0006513928849212874,
|
| 1499 |
+
"loss": 0.1628,
|
| 1500 |
+
"step": 213
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.18337617823479005,
|
| 1504 |
+
"grad_norm": 0.02975154109299183,
|
| 1505 |
+
"learning_rate": 0.0006482707874877854,
|
| 1506 |
+
"loss": 0.1671,
|
| 1507 |
+
"step": 214
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.1842330762639246,
|
| 1511 |
+
"grad_norm": 0.016304058954119682,
|
| 1512 |
+
"learning_rate": 0.0006451423386272311,
|
| 1513 |
+
"loss": 0.1681,
|
| 1514 |
+
"step": 215
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.18508997429305912,
|
| 1518 |
+
"grad_norm": 0.018919240683317184,
|
| 1519 |
+
"learning_rate": 0.0006420076723519614,
|
| 1520 |
+
"loss": 0.1632,
|
| 1521 |
+
"step": 216
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.18594687232219365,
|
| 1525 |
+
"grad_norm": 0.017546426504850388,
|
| 1526 |
+
"learning_rate": 0.0006388669229406462,
|
| 1527 |
+
"loss": 0.1626,
|
| 1528 |
+
"step": 217
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.1868037703513282,
|
| 1532 |
+
"grad_norm": 0.018862828612327576,
|
| 1533 |
+
"learning_rate": 0.0006357202249325371,
|
| 1534 |
+
"loss": 0.1779,
|
| 1535 |
+
"step": 218
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.18766066838046272,
|
| 1539 |
+
"grad_norm": 0.01685691997408867,
|
| 1540 |
+
"learning_rate": 0.000632567713121704,
|
| 1541 |
+
"loss": 0.1736,
|
| 1542 |
+
"step": 219
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.18851756640959727,
|
| 1546 |
+
"grad_norm": 0.017843585461378098,
|
| 1547 |
+
"learning_rate": 0.0006294095225512603,
|
| 1548 |
+
"loss": 0.1685,
|
| 1549 |
+
"step": 220
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.1893744644387318,
|
| 1553 |
+
"grad_norm": 0.019433507695794106,
|
| 1554 |
+
"learning_rate": 0.000626245788507579,
|
| 1555 |
+
"loss": 0.1601,
|
| 1556 |
+
"step": 221
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.19023136246786632,
|
| 1560 |
+
"grad_norm": 0.017471810802817345,
|
| 1561 |
+
"learning_rate": 0.0006230766465144965,
|
| 1562 |
+
"loss": 0.1626,
|
| 1563 |
+
"step": 222
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.19108826049700087,
|
| 1567 |
+
"grad_norm": 0.021109052002429962,
|
| 1568 |
+
"learning_rate": 0.0006199022323275083,
|
| 1569 |
+
"loss": 0.1608,
|
| 1570 |
+
"step": 223
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.1919451585261354,
|
| 1574 |
+
"grad_norm": 0.017889205366373062,
|
| 1575 |
+
"learning_rate": 0.0006167226819279528,
|
| 1576 |
+
"loss": 0.1666,
|
| 1577 |
+
"step": 224
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.1928020565552699,
|
| 1581 |
+
"grad_norm": 0.01807485520839691,
|
| 1582 |
+
"learning_rate": 0.0006135381315171866,
|
| 1583 |
+
"loss": 0.1717,
|
| 1584 |
+
"step": 225
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.19365895458440446,
|
| 1588 |
+
"grad_norm": 0.01634569652378559,
|
| 1589 |
+
"learning_rate": 0.0006103487175107507,
|
| 1590 |
+
"loss": 0.1698,
|
| 1591 |
+
"step": 226
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.194515852613539,
|
| 1595 |
+
"grad_norm": 0.019904734566807747,
|
| 1596 |
+
"learning_rate": 0.0006071545765325253,
|
| 1597 |
+
"loss": 0.1638,
|
| 1598 |
+
"step": 227
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.1953727506426735,
|
| 1602 |
+
"grad_norm": 0.020387910306453705,
|
| 1603 |
+
"learning_rate": 0.0006039558454088796,
|
| 1604 |
+
"loss": 0.1689,
|
| 1605 |
+
"step": 228
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.19622964867180806,
|
| 1609 |
+
"grad_norm": 0.018753819167613983,
|
| 1610 |
+
"learning_rate": 0.0006007526611628086,
|
| 1611 |
+
"loss": 0.1617,
|
| 1612 |
+
"step": 229
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.19708654670094258,
|
| 1616 |
+
"grad_norm": 0.015613901428878307,
|
| 1617 |
+
"learning_rate": 0.0005975451610080642,
|
| 1618 |
+
"loss": 0.1563,
|
| 1619 |
+
"step": 230
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.19794344473007713,
|
| 1623 |
+
"grad_norm": 0.018909510225057602,
|
| 1624 |
+
"learning_rate": 0.0005943334823432777,
|
| 1625 |
+
"loss": 0.1633,
|
| 1626 |
+
"step": 231
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.19880034275921166,
|
| 1630 |
+
"grad_norm": 0.01886546052992344,
|
| 1631 |
+
"learning_rate": 0.0005911177627460738,
|
| 1632 |
+
"loss": 0.1579,
|
| 1633 |
+
"step": 232
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.19965724078834618,
|
| 1637 |
+
"grad_norm": 0.014775911346077919,
|
| 1638 |
+
"learning_rate": 0.0005878981399671774,
|
| 1639 |
+
"loss": 0.1632,
|
| 1640 |
+
"step": 233
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.20051413881748073,
|
| 1644 |
+
"grad_norm": 0.01664350926876068,
|
| 1645 |
+
"learning_rate": 0.0005846747519245122,
|
| 1646 |
+
"loss": 0.1647,
|
| 1647 |
+
"step": 234
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.20137103684661525,
|
| 1651 |
+
"grad_norm": 0.015615535899996758,
|
| 1652 |
+
"learning_rate": 0.0005814477366972944,
|
| 1653 |
+
"loss": 0.1492,
|
| 1654 |
+
"step": 235
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.20222793487574978,
|
| 1658 |
+
"grad_norm": 0.020716892555356026,
|
| 1659 |
+
"learning_rate": 0.0005782172325201155,
|
| 1660 |
+
"loss": 0.166,
|
| 1661 |
+
"step": 236
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.20308483290488433,
|
| 1665 |
+
"grad_norm": 0.016642222180962563,
|
| 1666 |
+
"learning_rate": 0.0005749833777770225,
|
| 1667 |
+
"loss": 0.1598,
|
| 1668 |
+
"step": 237
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.20394173093401885,
|
| 1672 |
+
"grad_norm": 0.02147907204926014,
|
| 1673 |
+
"learning_rate": 0.0005717463109955896,
|
| 1674 |
+
"loss": 0.1604,
|
| 1675 |
+
"step": 238
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.20479862896315337,
|
| 1679 |
+
"grad_norm": 0.019267620518803596,
|
| 1680 |
+
"learning_rate": 0.0005685061708409841,
|
| 1681 |
+
"loss": 0.1558,
|
| 1682 |
+
"step": 239
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.20565552699228792,
|
| 1686 |
+
"grad_norm": 0.016083069145679474,
|
| 1687 |
+
"learning_rate": 0.000565263096110026,
|
| 1688 |
+
"loss": 0.1593,
|
| 1689 |
+
"step": 240
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.20651242502142245,
|
| 1693 |
+
"grad_norm": 0.023486295714974403,
|
| 1694 |
+
"learning_rate": 0.0005620172257252427,
|
| 1695 |
+
"loss": 0.1461,
|
| 1696 |
+
"step": 241
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.207369323050557,
|
| 1700 |
+
"grad_norm": 0.018065959215164185,
|
| 1701 |
+
"learning_rate": 0.0005587686987289189,
|
| 1702 |
+
"loss": 0.1673,
|
| 1703 |
+
"step": 242
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.20822622107969152,
|
| 1707 |
+
"grad_norm": 0.02077527344226837,
|
| 1708 |
+
"learning_rate": 0.0005555176542771388,
|
| 1709 |
+
"loss": 0.1617,
|
| 1710 |
+
"step": 243
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.20908311910882604,
|
| 1714 |
+
"grad_norm": 0.01682773232460022,
|
| 1715 |
+
"learning_rate": 0.0005522642316338268,
|
| 1716 |
+
"loss": 0.1626,
|
| 1717 |
+
"step": 244
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.2099400171379606,
|
| 1721 |
+
"grad_norm": 0.01779726706445217,
|
| 1722 |
+
"learning_rate": 0.0005490085701647804,
|
| 1723 |
+
"loss": 0.1585,
|
| 1724 |
+
"step": 245
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.21079691516709512,
|
| 1728 |
+
"grad_norm": 0.01758912205696106,
|
| 1729 |
+
"learning_rate": 0.0005457508093317013,
|
| 1730 |
+
"loss": 0.1642,
|
| 1731 |
+
"step": 246
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.21165381319622964,
|
| 1735 |
+
"grad_norm": 0.01853191666305065,
|
| 1736 |
+
"learning_rate": 0.0005424910886862209,
|
| 1737 |
+
"loss": 0.1729,
|
| 1738 |
+
"step": 247
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.2125107112253642,
|
| 1742 |
+
"grad_norm": 0.015318380668759346,
|
| 1743 |
+
"learning_rate": 0.0005392295478639225,
|
| 1744 |
+
"loss": 0.1519,
|
| 1745 |
+
"step": 248
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.2133676092544987,
|
| 1749 |
+
"grad_norm": 0.016783015802502632,
|
| 1750 |
+
"learning_rate": 0.0005359663265783598,
|
| 1751 |
+
"loss": 0.1534,
|
| 1752 |
+
"step": 249
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.21422450728363324,
|
| 1756 |
+
"grad_norm": 0.019508186727762222,
|
| 1757 |
+
"learning_rate": 0.0005327015646150716,
|
| 1758 |
+
"loss": 0.1573,
|
| 1759 |
+
"step": 250
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.2150814053127678,
|
| 1763 |
+
"grad_norm": 0.017221014946699142,
|
| 1764 |
+
"learning_rate": 0.0005294354018255945,
|
| 1765 |
+
"loss": 0.168,
|
| 1766 |
+
"step": 251
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.2159383033419023,
|
| 1770 |
+
"grad_norm": 0.015044581145048141,
|
| 1771 |
+
"learning_rate": 0.000526167978121472,
|
| 1772 |
+
"loss": 0.168,
|
| 1773 |
+
"step": 252
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.21679520137103683,
|
| 1777 |
+
"grad_norm": 0.016873784363269806,
|
| 1778 |
+
"learning_rate": 0.0005228994334682604,
|
| 1779 |
+
"loss": 0.1585,
|
| 1780 |
+
"step": 253
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.21765209940017138,
|
| 1784 |
+
"grad_norm": 0.017317088320851326,
|
| 1785 |
+
"learning_rate": 0.0005196299078795343,
|
| 1786 |
+
"loss": 0.1556,
|
| 1787 |
+
"step": 254
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.2185089974293059,
|
| 1791 |
+
"grad_norm": 0.017208045348525047,
|
| 1792 |
+
"learning_rate": 0.0005163595414108881,
|
| 1793 |
+
"loss": 0.1551,
|
| 1794 |
+
"step": 255
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.21936589545844046,
|
| 1798 |
+
"grad_norm": 0.015319808386266232,
|
| 1799 |
+
"learning_rate": 0.0005130884741539367,
|
| 1800 |
+
"loss": 0.1491,
|
| 1801 |
+
"step": 256
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.22022279348757498,
|
| 1805 |
+
"grad_norm": 0.020643778145313263,
|
| 1806 |
+
"learning_rate": 0.0005098168462303141,
|
| 1807 |
+
"loss": 0.1629,
|
| 1808 |
+
"step": 257
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.2210796915167095,
|
| 1812 |
+
"grad_norm": 0.01764957793056965,
|
| 1813 |
+
"learning_rate": 0.0005065447977856722,
|
| 1814 |
+
"loss": 0.16,
|
| 1815 |
+
"step": 258
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.22193658954584405,
|
| 1819 |
+
"grad_norm": 0.014267503283917904,
|
| 1820 |
+
"learning_rate": 0.0005032724689836759,
|
| 1821 |
+
"loss": 0.1524,
|
| 1822 |
+
"step": 259
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.22279348757497858,
|
| 1826 |
+
"grad_norm": 0.016303062438964844,
|
| 1827 |
+
"learning_rate": 0.0005,
|
| 1828 |
+
"loss": 0.1544,
|
| 1829 |
+
"step": 260
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.2236503856041131,
|
| 1833 |
+
"grad_norm": 0.021528156474232674,
|
| 1834 |
+
"learning_rate": 0.0004967275310163241,
|
| 1835 |
+
"loss": 0.155,
|
| 1836 |
+
"step": 261
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.22450728363324765,
|
| 1840 |
+
"grad_norm": 0.017993303015828133,
|
| 1841 |
+
"learning_rate": 0.0004934552022143279,
|
| 1842 |
+
"loss": 0.1646,
|
| 1843 |
+
"step": 262
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.22536418166238217,
|
| 1847 |
+
"grad_norm": 0.018955664709210396,
|
| 1848 |
+
"learning_rate": 0.0004901831537696859,
|
| 1849 |
+
"loss": 0.1564,
|
| 1850 |
+
"step": 263
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.2262210796915167,
|
| 1854 |
+
"grad_norm": 0.01549836527556181,
|
| 1855 |
+
"learning_rate": 0.0004869115258460635,
|
| 1856 |
+
"loss": 0.1605,
|
| 1857 |
+
"step": 264
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.22707797772065125,
|
| 1861 |
+
"grad_norm": 0.019947407767176628,
|
| 1862 |
+
"learning_rate": 0.00048364045858911197,
|
| 1863 |
+
"loss": 0.1586,
|
| 1864 |
+
"step": 265
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.22793487574978577,
|
| 1868 |
+
"grad_norm": 0.022805072367191315,
|
| 1869 |
+
"learning_rate": 0.00048037009212046586,
|
| 1870 |
+
"loss": 0.1585,
|
| 1871 |
+
"step": 266
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.22879177377892032,
|
| 1875 |
+
"grad_norm": 0.014930406585335732,
|
| 1876 |
+
"learning_rate": 0.0004771005665317397,
|
| 1877 |
+
"loss": 0.1541,
|
| 1878 |
+
"step": 267
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.22964867180805484,
|
| 1882 |
+
"grad_norm": 0.017478667199611664,
|
| 1883 |
+
"learning_rate": 0.0004738320218785281,
|
| 1884 |
+
"loss": 0.1652,
|
| 1885 |
+
"step": 268
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.23050556983718937,
|
| 1889 |
+
"grad_norm": 0.016168439760804176,
|
| 1890 |
+
"learning_rate": 0.00047056459817440544,
|
| 1891 |
+
"loss": 0.1558,
|
| 1892 |
+
"step": 269
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.23136246786632392,
|
| 1896 |
+
"grad_norm": 0.018146967515349388,
|
| 1897 |
+
"learning_rate": 0.00046729843538492847,
|
| 1898 |
+
"loss": 0.1589,
|
| 1899 |
+
"step": 270
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.23221936589545844,
|
| 1903 |
+
"grad_norm": 0.021017901599407196,
|
| 1904 |
+
"learning_rate": 0.00046403367342164026,
|
| 1905 |
+
"loss": 0.158,
|
| 1906 |
+
"step": 271
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.23307626392459296,
|
| 1910 |
+
"grad_norm": 0.0199353639036417,
|
| 1911 |
+
"learning_rate": 0.0004607704521360776,
|
| 1912 |
+
"loss": 0.159,
|
| 1913 |
+
"step": 272
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.23393316195372751,
|
| 1917 |
+
"grad_norm": 0.0214347206056118,
|
| 1918 |
+
"learning_rate": 0.0004575089113137792,
|
| 1919 |
+
"loss": 0.1594,
|
| 1920 |
+
"step": 273
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.23479005998286204,
|
| 1924 |
+
"grad_norm": 0.01558151189237833,
|
| 1925 |
+
"learning_rate": 0.00045424919066829885,
|
| 1926 |
+
"loss": 0.1595,
|
| 1927 |
+
"step": 274
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.23564695801199656,
|
| 1931 |
+
"grad_norm": 0.01718144491314888,
|
| 1932 |
+
"learning_rate": 0.0004509914298352197,
|
| 1933 |
+
"loss": 0.1636,
|
| 1934 |
+
"step": 275
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.2365038560411311,
|
| 1938 |
+
"grad_norm": 0.0169569943100214,
|
| 1939 |
+
"learning_rate": 0.00044773576836617336,
|
| 1940 |
+
"loss": 0.1581,
|
| 1941 |
+
"step": 276
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.23736075407026563,
|
| 1945 |
+
"grad_norm": 0.01732802391052246,
|
| 1946 |
+
"learning_rate": 0.0004444823457228612,
|
| 1947 |
+
"loss": 0.1544,
|
| 1948 |
+
"step": 277
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.23821765209940018,
|
| 1952 |
+
"grad_norm": 0.01665407046675682,
|
| 1953 |
+
"learning_rate": 0.00044123130127108126,
|
| 1954 |
+
"loss": 0.1607,
|
| 1955 |
+
"step": 278
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.2390745501285347,
|
| 1959 |
+
"grad_norm": 0.0163556020706892,
|
| 1960 |
+
"learning_rate": 0.0004379827742747575,
|
| 1961 |
+
"loss": 0.1544,
|
| 1962 |
+
"step": 279
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.23993144815766923,
|
| 1966 |
+
"grad_norm": 0.01844378188252449,
|
| 1967 |
+
"learning_rate": 0.00043473690388997434,
|
| 1968 |
+
"loss": 0.1552,
|
| 1969 |
+
"step": 280
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.24078834618680378,
|
| 1973 |
+
"grad_norm": 0.01735353097319603,
|
| 1974 |
+
"learning_rate": 0.0004314938291590161,
|
| 1975 |
+
"loss": 0.1592,
|
| 1976 |
+
"step": 281
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.2416452442159383,
|
| 1980 |
+
"grad_norm": 0.01745842583477497,
|
| 1981 |
+
"learning_rate": 0.0004282536890044104,
|
| 1982 |
+
"loss": 0.1548,
|
| 1983 |
+
"step": 282
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.24250214224507283,
|
| 1987 |
+
"grad_norm": 0.015916811302304268,
|
| 1988 |
+
"learning_rate": 0.0004250166222229774,
|
| 1989 |
+
"loss": 0.1512,
|
| 1990 |
+
"step": 283
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.24335904027420738,
|
| 1994 |
+
"grad_norm": 0.015339579433202744,
|
| 1995 |
+
"learning_rate": 0.0004217827674798845,
|
| 1996 |
+
"loss": 0.1553,
|
| 1997 |
+
"step": 284
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.2442159383033419,
|
| 2001 |
+
"grad_norm": 0.01986338384449482,
|
| 2002 |
+
"learning_rate": 0.0004185522633027057,
|
| 2003 |
+
"loss": 0.163,
|
| 2004 |
+
"step": 285
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.24507283633247642,
|
| 2008 |
+
"grad_norm": 0.018619602546095848,
|
| 2009 |
+
"learning_rate": 0.0004153252480754877,
|
| 2010 |
+
"loss": 0.1526,
|
| 2011 |
+
"step": 286
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.24592973436161097,
|
| 2015 |
+
"grad_norm": 0.016342243179678917,
|
| 2016 |
+
"learning_rate": 0.00041210186003282274,
|
| 2017 |
+
"loss": 0.157,
|
| 2018 |
+
"step": 287
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.2467866323907455,
|
| 2022 |
+
"grad_norm": 0.016473161056637764,
|
| 2023 |
+
"learning_rate": 0.00040888223725392626,
|
| 2024 |
+
"loss": 0.1638,
|
| 2025 |
+
"step": 288
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.24764353041988005,
|
| 2029 |
+
"grad_norm": 0.015475657768547535,
|
| 2030 |
+
"learning_rate": 0.00040566651765672245,
|
| 2031 |
+
"loss": 0.1566,
|
| 2032 |
+
"step": 289
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.24850042844901457,
|
| 2036 |
+
"grad_norm": 0.019614532589912415,
|
| 2037 |
+
"learning_rate": 0.00040245483899193594,
|
| 2038 |
+
"loss": 0.1563,
|
| 2039 |
+
"step": 290
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.2493573264781491,
|
| 2043 |
+
"grad_norm": 0.01654512993991375,
|
| 2044 |
+
"learning_rate": 0.00039924733883719147,
|
| 2045 |
+
"loss": 0.155,
|
| 2046 |
+
"step": 291
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.25021422450728364,
|
| 2050 |
+
"grad_norm": 0.01950278878211975,
|
| 2051 |
+
"learning_rate": 0.0003960441545911204,
|
| 2052 |
+
"loss": 0.1562,
|
| 2053 |
+
"step": 292
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.25107112253641817,
|
| 2057 |
+
"grad_norm": 0.014652718789875507,
|
| 2058 |
+
"learning_rate": 0.0003928454234674747,
|
| 2059 |
+
"loss": 0.1524,
|
| 2060 |
+
"step": 293
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.2519280205655527,
|
| 2064 |
+
"grad_norm": 0.014939884655177593,
|
| 2065 |
+
"learning_rate": 0.0003896512824892495,
|
| 2066 |
+
"loss": 0.1541,
|
| 2067 |
+
"step": 294
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.2527849185946872,
|
| 2071 |
+
"grad_norm": 0.0260702446103096,
|
| 2072 |
+
"learning_rate": 0.00038646186848281344,
|
| 2073 |
+
"loss": 0.1603,
|
| 2074 |
+
"step": 295
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.2536418166238218,
|
| 2078 |
+
"grad_norm": 0.022277580574154854,
|
| 2079 |
+
"learning_rate": 0.00038327731807204744,
|
| 2080 |
+
"loss": 0.1559,
|
| 2081 |
+
"step": 296
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.2544987146529563,
|
| 2085 |
+
"grad_norm": 0.017295408993959427,
|
| 2086 |
+
"learning_rate": 0.0003800977676724919,
|
| 2087 |
+
"loss": 0.1641,
|
| 2088 |
+
"step": 297
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.25535561268209084,
|
| 2092 |
+
"grad_norm": 0.015900392085313797,
|
| 2093 |
+
"learning_rate": 0.0003769233534855035,
|
| 2094 |
+
"loss": 0.1569,
|
| 2095 |
+
"step": 298
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.25621251071122536,
|
| 2099 |
+
"grad_norm": 0.02040684036910534,
|
| 2100 |
+
"learning_rate": 0.00037375421149242103,
|
| 2101 |
+
"loss": 0.1529,
|
| 2102 |
+
"step": 299
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.2570694087403599,
|
| 2106 |
+
"grad_norm": 0.020803892984986305,
|
| 2107 |
+
"learning_rate": 0.0003705904774487396,
|
| 2108 |
+
"loss": 0.1546,
|
| 2109 |
+
"step": 300
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.2579263067694944,
|
| 2113 |
+
"grad_norm": 0.017124850302934647,
|
| 2114 |
+
"learning_rate": 0.0003674322868782959,
|
| 2115 |
+
"loss": 0.1499,
|
| 2116 |
+
"step": 301
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.258783204798629,
|
| 2120 |
+
"grad_norm": 0.015450037084519863,
|
| 2121 |
+
"learning_rate": 0.0003642797750674629,
|
| 2122 |
+
"loss": 0.1485,
|
| 2123 |
+
"step": 302
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.2596401028277635,
|
| 2127 |
+
"grad_norm": 0.017722809687256813,
|
| 2128 |
+
"learning_rate": 0.00036113307705935393,
|
| 2129 |
+
"loss": 0.158,
|
| 2130 |
+
"step": 303
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.26049700085689803,
|
| 2134 |
+
"grad_norm": 0.016334377229213715,
|
| 2135 |
+
"learning_rate": 0.0003579923276480387,
|
| 2136 |
+
"loss": 0.1583,
|
| 2137 |
+
"step": 304
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.26135389888603255,
|
| 2141 |
+
"grad_norm": 0.014489492401480675,
|
| 2142 |
+
"learning_rate": 0.0003548576613727689,
|
| 2143 |
+
"loss": 0.1493,
|
| 2144 |
+
"step": 305
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.2622107969151671,
|
| 2148 |
+
"grad_norm": 0.022003574296832085,
|
| 2149 |
+
"learning_rate": 0.0003517292125122146,
|
| 2150 |
+
"loss": 0.159,
|
| 2151 |
+
"step": 306
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.26306769494430166,
|
| 2155 |
+
"grad_norm": 0.01759640872478485,
|
| 2156 |
+
"learning_rate": 0.0003486071150787128,
|
| 2157 |
+
"loss": 0.1485,
|
| 2158 |
+
"step": 307
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.2639245929734362,
|
| 2162 |
+
"grad_norm": 0.017299624159932137,
|
| 2163 |
+
"learning_rate": 0.00034549150281252633,
|
| 2164 |
+
"loss": 0.1552,
|
| 2165 |
+
"step": 308
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.2647814910025707,
|
| 2169 |
+
"grad_norm": 0.015600617974996567,
|
| 2170 |
+
"learning_rate": 0.0003423825091761153,
|
| 2171 |
+
"loss": 0.1535,
|
| 2172 |
+
"step": 309
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.2656383890317052,
|
| 2176 |
+
"grad_norm": 0.016550280153751373,
|
| 2177 |
+
"learning_rate": 0.0003392802673484193,
|
| 2178 |
+
"loss": 0.1513,
|
| 2179 |
+
"step": 310
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.26649528706083975,
|
| 2183 |
+
"grad_norm": 0.014733058400452137,
|
| 2184 |
+
"learning_rate": 0.0003361849102191533,
|
| 2185 |
+
"loss": 0.1542,
|
| 2186 |
+
"step": 311
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.26735218508997427,
|
| 2190 |
+
"grad_norm": 0.015453443862497807,
|
| 2191 |
+
"learning_rate": 0.00033309657038311456,
|
| 2192 |
+
"loss": 0.1567,
|
| 2193 |
+
"step": 312
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.26820908311910885,
|
| 2197 |
+
"grad_norm": 0.01371778268367052,
|
| 2198 |
+
"learning_rate": 0.00033001538013450283,
|
| 2199 |
+
"loss": 0.1545,
|
| 2200 |
+
"step": 313
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.26906598114824337,
|
| 2204 |
+
"grad_norm": 0.014121933840215206,
|
| 2205 |
+
"learning_rate": 0.0003269414714612534,
|
| 2206 |
+
"loss": 0.1579,
|
| 2207 |
+
"step": 314
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.2699228791773779,
|
| 2211 |
+
"grad_norm": 0.012276604771614075,
|
| 2212 |
+
"learning_rate": 0.00032387497603938325,
|
| 2213 |
+
"loss": 0.1493,
|
| 2214 |
+
"step": 315
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.2707797772065124,
|
| 2218 |
+
"grad_norm": 0.017444469034671783,
|
| 2219 |
+
"learning_rate": 0.00032081602522734986,
|
| 2220 |
+
"loss": 0.1535,
|
| 2221 |
+
"step": 316
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.27163667523564694,
|
| 2225 |
+
"grad_norm": 0.01802118867635727,
|
| 2226 |
+
"learning_rate": 0.0003177647500604252,
|
| 2227 |
+
"loss": 0.1556,
|
| 2228 |
+
"step": 317
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 0.27249357326478146,
|
| 2232 |
+
"grad_norm": 0.014814218506217003,
|
| 2233 |
+
"learning_rate": 0.00031472128124508187,
|
| 2234 |
+
"loss": 0.1534,
|
| 2235 |
+
"step": 318
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 0.27335047129391604,
|
| 2239 |
+
"grad_norm": 0.02104412205517292,
|
| 2240 |
+
"learning_rate": 0.00031168574915339467,
|
| 2241 |
+
"loss": 0.1572,
|
| 2242 |
+
"step": 319
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.27420736932305056,
|
| 2246 |
+
"grad_norm": 0.014852729625999928,
|
| 2247 |
+
"learning_rate": 0.0003086582838174551,
|
| 2248 |
+
"loss": 0.1509,
|
| 2249 |
+
"step": 320
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 0.2750642673521851,
|
| 2253 |
+
"grad_norm": 0.013851814903318882,
|
| 2254 |
+
"learning_rate": 0.0003056390149238022,
|
| 2255 |
+
"loss": 0.1557,
|
| 2256 |
+
"step": 321
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 0.2759211653813196,
|
| 2260 |
+
"grad_norm": 0.014435027725994587,
|
| 2261 |
+
"learning_rate": 0.00030262807180786645,
|
| 2262 |
+
"loss": 0.1516,
|
| 2263 |
+
"step": 322
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 0.27677806341045413,
|
| 2267 |
+
"grad_norm": 0.014334925450384617,
|
| 2268 |
+
"learning_rate": 0.00029962558344842963,
|
| 2269 |
+
"loss": 0.1542,
|
| 2270 |
+
"step": 323
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.2776349614395887,
|
| 2274 |
+
"grad_norm": 0.017674414440989494,
|
| 2275 |
+
"learning_rate": 0.0002966316784621,
|
| 2276 |
+
"loss": 0.1525,
|
| 2277 |
+
"step": 324
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 0.27849185946872324,
|
| 2281 |
+
"grad_norm": 0.019711866974830627,
|
| 2282 |
+
"learning_rate": 0.0002936464850978027,
|
| 2283 |
+
"loss": 0.1486,
|
| 2284 |
+
"step": 325
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 0.27934875749785776,
|
| 2288 |
+
"grad_norm": 0.015335663221776485,
|
| 2289 |
+
"learning_rate": 0.0002906701312312861,
|
| 2290 |
+
"loss": 0.1508,
|
| 2291 |
+
"step": 326
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 0.2802056555269923,
|
| 2295 |
+
"grad_norm": 0.013025953434407711,
|
| 2296 |
+
"learning_rate": 0.00028770274435964356,
|
| 2297 |
+
"loss": 0.1499,
|
| 2298 |
+
"step": 327
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 0.2810625535561268,
|
| 2302 |
+
"grad_norm": 0.01768515445291996,
|
| 2303 |
+
"learning_rate": 0.0002847444515958523,
|
| 2304 |
+
"loss": 0.1527,
|
| 2305 |
+
"step": 328
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 0.2819194515852613,
|
| 2309 |
+
"grad_norm": 0.015223097056150436,
|
| 2310 |
+
"learning_rate": 0.0002817953796633289,
|
| 2311 |
+
"loss": 0.1549,
|
| 2312 |
+
"step": 329
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 0.2827763496143959,
|
| 2316 |
+
"grad_norm": 0.013024591840803623,
|
| 2317 |
+
"learning_rate": 0.00027885565489049947,
|
| 2318 |
+
"loss": 0.1509,
|
| 2319 |
+
"step": 330
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.28363324764353043,
|
| 2323 |
+
"grad_norm": 0.014075031504034996,
|
| 2324 |
+
"learning_rate": 0.0002759254032053888,
|
| 2325 |
+
"loss": 0.1558,
|
| 2326 |
+
"step": 331
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 0.28449014567266495,
|
| 2330 |
+
"grad_norm": 0.013003628700971603,
|
| 2331 |
+
"learning_rate": 0.00027300475013022663,
|
| 2332 |
+
"loss": 0.15,
|
| 2333 |
+
"step": 332
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 0.2853470437017995,
|
| 2337 |
+
"grad_norm": 0.014489670284092426,
|
| 2338 |
+
"learning_rate": 0.0002700938207760701,
|
| 2339 |
+
"loss": 0.1476,
|
| 2340 |
+
"step": 333
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 0.286203941730934,
|
| 2344 |
+
"grad_norm": 0.012941529043018818,
|
| 2345 |
+
"learning_rate": 0.0002671927398374443,
|
| 2346 |
+
"loss": 0.1408,
|
| 2347 |
+
"step": 334
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 0.2870608397600686,
|
| 2351 |
+
"grad_norm": 0.012417233549058437,
|
| 2352 |
+
"learning_rate": 0.00026430163158700117,
|
| 2353 |
+
"loss": 0.1485,
|
| 2354 |
+
"step": 335
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 0.2879177377892031,
|
| 2358 |
+
"grad_norm": 0.01331823505461216,
|
| 2359 |
+
"learning_rate": 0.00026142061987019576,
|
| 2360 |
+
"loss": 0.1578,
|
| 2361 |
+
"step": 336
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 0.2887746358183376,
|
| 2365 |
+
"grad_norm": 0.021479349583387375,
|
| 2366 |
+
"learning_rate": 0.0002585498280999815,
|
| 2367 |
+
"loss": 0.1544,
|
| 2368 |
+
"step": 337
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 0.28963153384747214,
|
| 2372 |
+
"grad_norm": 0.014903098344802856,
|
| 2373 |
+
"learning_rate": 0.0002556893792515227,
|
| 2374 |
+
"loss": 0.1561,
|
| 2375 |
+
"step": 338
|
| 2376 |
+
},
|
| 2377 |
+
{
|
| 2378 |
+
"epoch": 0.29048843187660667,
|
| 2379 |
+
"grad_norm": 0.013264741748571396,
|
| 2380 |
+
"learning_rate": 0.00025283939585692784,
|
| 2381 |
+
"loss": 0.1599,
|
| 2382 |
+
"step": 339
|
| 2383 |
+
},
|
| 2384 |
+
{
|
| 2385 |
+
"epoch": 0.2913453299057412,
|
| 2386 |
+
"grad_norm": 0.0137154096737504,
|
| 2387 |
+
"learning_rate": 0.0002500000000000001,
|
| 2388 |
+
"loss": 0.1526,
|
| 2389 |
+
"step": 340
|
| 2390 |
+
},
|
| 2391 |
+
{
|
| 2392 |
+
"epoch": 0.29220222793487577,
|
| 2393 |
+
"grad_norm": 0.01252024993300438,
|
| 2394 |
+
"learning_rate": 0.0002471713133110078,
|
| 2395 |
+
"loss": 0.1421,
|
| 2396 |
+
"step": 341
|
| 2397 |
+
},
|
| 2398 |
+
{
|
| 2399 |
+
"epoch": 0.2930591259640103,
|
| 2400 |
+
"grad_norm": 0.01306652370840311,
|
| 2401 |
+
"learning_rate": 0.00024435345696147403,
|
| 2402 |
+
"loss": 0.1373,
|
| 2403 |
+
"step": 342
|
| 2404 |
+
},
|
| 2405 |
+
{
|
| 2406 |
+
"epoch": 0.2939160239931448,
|
| 2407 |
+
"grad_norm": 0.015782173722982407,
|
| 2408 |
+
"learning_rate": 0.00024154655165898627,
|
| 2409 |
+
"loss": 0.149,
|
| 2410 |
+
"step": 343
|
| 2411 |
+
},
|
| 2412 |
+
{
|
| 2413 |
+
"epoch": 0.29477292202227934,
|
| 2414 |
+
"grad_norm": 0.012606021016836166,
|
| 2415 |
+
"learning_rate": 0.00023875071764202561,
|
| 2416 |
+
"loss": 0.153,
|
| 2417 |
+
"step": 344
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"epoch": 0.29562982005141386,
|
| 2421 |
+
"grad_norm": 0.013155002146959305,
|
| 2422 |
+
"learning_rate": 0.00023596607467481602,
|
| 2423 |
+
"loss": 0.1493,
|
| 2424 |
+
"step": 345
|
| 2425 |
+
},
|
| 2426 |
+
{
|
| 2427 |
+
"epoch": 0.29648671808054844,
|
| 2428 |
+
"grad_norm": 0.01664326898753643,
|
| 2429 |
+
"learning_rate": 0.00023319274204219425,
|
| 2430 |
+
"loss": 0.1566,
|
| 2431 |
+
"step": 346
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 0.29734361610968296,
|
| 2435 |
+
"grad_norm": 0.01268248911947012,
|
| 2436 |
+
"learning_rate": 0.00023043083854449987,
|
| 2437 |
+
"loss": 0.1467,
|
| 2438 |
+
"step": 347
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 0.2982005141388175,
|
| 2442 |
+
"grad_norm": 0.012720319442451,
|
| 2443 |
+
"learning_rate": 0.00022768048249248646,
|
| 2444 |
+
"loss": 0.1528,
|
| 2445 |
+
"step": 348
|
| 2446 |
+
},
|
| 2447 |
+
{
|
| 2448 |
+
"epoch": 0.299057412167952,
|
| 2449 |
+
"grad_norm": 0.012999819591641426,
|
| 2450 |
+
"learning_rate": 0.00022494179170225333,
|
| 2451 |
+
"loss": 0.1549,
|
| 2452 |
+
"step": 349
|
| 2453 |
+
},
|
| 2454 |
+
{
|
| 2455 |
+
"epoch": 0.29991431019708653,
|
| 2456 |
+
"grad_norm": 0.013431290164589882,
|
| 2457 |
+
"learning_rate": 0.00022221488349019903,
|
| 2458 |
+
"loss": 0.1525,
|
| 2459 |
+
"step": 350
|
| 2460 |
+
},
|
| 2461 |
+
{
|
| 2462 |
+
"epoch": 0.30077120822622105,
|
| 2463 |
+
"grad_norm": 0.01783970557153225,
|
| 2464 |
+
"learning_rate": 0.0002194998746679952,
|
| 2465 |
+
"loss": 0.1527,
|
| 2466 |
+
"step": 351
|
| 2467 |
+
},
|
| 2468 |
+
{
|
| 2469 |
+
"epoch": 0.30162810625535563,
|
| 2470 |
+
"grad_norm": 0.021057991310954094,
|
| 2471 |
+
"learning_rate": 0.0002167968815375837,
|
| 2472 |
+
"loss": 0.1538,
|
| 2473 |
+
"step": 352
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 0.30248500428449016,
|
| 2477 |
+
"grad_norm": 0.01601223647594452,
|
| 2478 |
+
"learning_rate": 0.00021410601988619394,
|
| 2479 |
+
"loss": 0.1525,
|
| 2480 |
+
"step": 353
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 0.3033419023136247,
|
| 2484 |
+
"grad_norm": 0.012890568003058434,
|
| 2485 |
+
"learning_rate": 0.00021142740498138323,
|
| 2486 |
+
"loss": 0.1488,
|
| 2487 |
+
"step": 354
|
| 2488 |
+
},
|
| 2489 |
+
{
|
| 2490 |
+
"epoch": 0.3041988003427592,
|
| 2491 |
+
"grad_norm": 0.013061465695500374,
|
| 2492 |
+
"learning_rate": 0.000208761151566099,
|
| 2493 |
+
"loss": 0.1498,
|
| 2494 |
+
"step": 355
|
| 2495 |
+
},
|
| 2496 |
+
{
|
| 2497 |
+
"epoch": 0.3050556983718937,
|
| 2498 |
+
"grad_norm": 0.013169737532734871,
|
| 2499 |
+
"learning_rate": 0.00020610737385376348,
|
| 2500 |
+
"loss": 0.15,
|
| 2501 |
+
"step": 356
|
| 2502 |
+
},
|
| 2503 |
+
{
|
| 2504 |
+
"epoch": 0.3059125964010283,
|
| 2505 |
+
"grad_norm": 0.012198954820632935,
|
| 2506 |
+
"learning_rate": 0.00020346618552338148,
|
| 2507 |
+
"loss": 0.1567,
|
| 2508 |
+
"step": 357
|
| 2509 |
+
},
|
| 2510 |
+
{
|
| 2511 |
+
"epoch": 0.3067694944301628,
|
| 2512 |
+
"grad_norm": 0.01644357666373253,
|
| 2513 |
+
"learning_rate": 0.00020083769971467047,
|
| 2514 |
+
"loss": 0.1532,
|
| 2515 |
+
"step": 358
|
| 2516 |
+
},
|
| 2517 |
+
{
|
| 2518 |
+
"epoch": 0.30762639245929735,
|
| 2519 |
+
"grad_norm": 0.014105524867773056,
|
| 2520 |
+
"learning_rate": 0.0001982220290232143,
|
| 2521 |
+
"loss": 0.1585,
|
| 2522 |
+
"step": 359
|
| 2523 |
+
},
|
| 2524 |
+
{
|
| 2525 |
+
"epoch": 0.30848329048843187,
|
| 2526 |
+
"grad_norm": 0.012292543426156044,
|
| 2527 |
+
"learning_rate": 0.00019561928549563967,
|
| 2528 |
+
"loss": 0.1525,
|
| 2529 |
+
"step": 360
|
| 2530 |
+
},
|
| 2531 |
+
{
|
| 2532 |
+
"epoch": 0.3093401885175664,
|
| 2533 |
+
"grad_norm": 0.01830083131790161,
|
| 2534 |
+
"learning_rate": 0.00019302958062481672,
|
| 2535 |
+
"loss": 0.1506,
|
| 2536 |
+
"step": 361
|
| 2537 |
+
},
|
| 2538 |
+
{
|
| 2539 |
+
"epoch": 0.3101970865467009,
|
| 2540 |
+
"grad_norm": 0.013068013824522495,
|
| 2541 |
+
"learning_rate": 0.00019045302534508295,
|
| 2542 |
+
"loss": 0.1563,
|
| 2543 |
+
"step": 362
|
| 2544 |
+
},
|
| 2545 |
+
{
|
| 2546 |
+
"epoch": 0.3110539845758355,
|
| 2547 |
+
"grad_norm": 0.012616029009222984,
|
| 2548 |
+
"learning_rate": 0.0001878897300274911,
|
| 2549 |
+
"loss": 0.1477,
|
| 2550 |
+
"step": 363
|
| 2551 |
+
},
|
| 2552 |
+
{
|
| 2553 |
+
"epoch": 0.31191088260497,
|
| 2554 |
+
"grad_norm": 0.015404236502945423,
|
| 2555 |
+
"learning_rate": 0.00018533980447508135,
|
| 2556 |
+
"loss": 0.1501,
|
| 2557 |
+
"step": 364
|
| 2558 |
+
},
|
| 2559 |
+
{
|
| 2560 |
+
"epoch": 0.31276778063410454,
|
| 2561 |
+
"grad_norm": 0.017586475238204002,
|
| 2562 |
+
"learning_rate": 0.00018280335791817732,
|
| 2563 |
+
"loss": 0.1493,
|
| 2564 |
+
"step": 365
|
| 2565 |
+
},
|
| 2566 |
+
{
|
| 2567 |
+
"epoch": 0.31362467866323906,
|
| 2568 |
+
"grad_norm": 0.01738077774643898,
|
| 2569 |
+
"learning_rate": 0.00018028049900970766,
|
| 2570 |
+
"loss": 0.1571,
|
| 2571 |
+
"step": 366
|
| 2572 |
+
},
|
| 2573 |
+
{
|
| 2574 |
+
"epoch": 0.3144815766923736,
|
| 2575 |
+
"grad_norm": 0.01235333550721407,
|
| 2576 |
+
"learning_rate": 0.0001777713358205514,
|
| 2577 |
+
"loss": 0.1528,
|
| 2578 |
+
"step": 367
|
| 2579 |
+
},
|
| 2580 |
+
{
|
| 2581 |
+
"epoch": 0.31533847472150817,
|
| 2582 |
+
"grad_norm": 0.01204688660800457,
|
| 2583 |
+
"learning_rate": 0.00017527597583490823,
|
| 2584 |
+
"loss": 0.1483,
|
| 2585 |
+
"step": 368
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 0.3161953727506427,
|
| 2589 |
+
"grad_norm": 0.012409962713718414,
|
| 2590 |
+
"learning_rate": 0.00017279452594569483,
|
| 2591 |
+
"loss": 0.1495,
|
| 2592 |
+
"step": 369
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 0.3170522707797772,
|
| 2596 |
+
"grad_norm": 0.010938968509435654,
|
| 2597 |
+
"learning_rate": 0.00017032709244996558,
|
| 2598 |
+
"loss": 0.149,
|
| 2599 |
+
"step": 370
|
| 2600 |
+
},
|
| 2601 |
+
{
|
| 2602 |
+
"epoch": 0.31790916880891174,
|
| 2603 |
+
"grad_norm": 0.012267996557056904,
|
| 2604 |
+
"learning_rate": 0.00016787378104435928,
|
| 2605 |
+
"loss": 0.154,
|
| 2606 |
+
"step": 371
|
| 2607 |
+
},
|
| 2608 |
+
{
|
| 2609 |
+
"epoch": 0.31876606683804626,
|
| 2610 |
+
"grad_norm": 0.012454659678041935,
|
| 2611 |
+
"learning_rate": 0.00016543469682057105,
|
| 2612 |
+
"loss": 0.1509,
|
| 2613 |
+
"step": 372
|
| 2614 |
+
},
|
| 2615 |
+
{
|
| 2616 |
+
"epoch": 0.3196229648671808,
|
| 2617 |
+
"grad_norm": 0.01519513875246048,
|
| 2618 |
+
"learning_rate": 0.00016300994426085103,
|
| 2619 |
+
"loss": 0.1539,
|
| 2620 |
+
"step": 373
|
| 2621 |
+
},
|
| 2622 |
+
{
|
| 2623 |
+
"epoch": 0.32047986289631536,
|
| 2624 |
+
"grad_norm": 0.013357303105294704,
|
| 2625 |
+
"learning_rate": 0.0001605996272335291,
|
| 2626 |
+
"loss": 0.1573,
|
| 2627 |
+
"step": 374
|
| 2628 |
+
},
|
| 2629 |
+
{
|
| 2630 |
+
"epoch": 0.3213367609254499,
|
| 2631 |
+
"grad_norm": 0.012079097330570221,
|
| 2632 |
+
"learning_rate": 0.00015820384898856434,
|
| 2633 |
+
"loss": 0.1555,
|
| 2634 |
+
"step": 375
|
| 2635 |
+
},
|
| 2636 |
+
{
|
| 2637 |
+
"epoch": 0.3221936589545844,
|
| 2638 |
+
"grad_norm": 0.012022243812680244,
|
| 2639 |
+
"learning_rate": 0.00015582271215312294,
|
| 2640 |
+
"loss": 0.1414,
|
| 2641 |
+
"step": 376
|
| 2642 |
+
},
|
| 2643 |
+
{
|
| 2644 |
+
"epoch": 0.32305055698371893,
|
| 2645 |
+
"grad_norm": 0.012276671826839447,
|
| 2646 |
+
"learning_rate": 0.00015345631872718213,
|
| 2647 |
+
"loss": 0.1457,
|
| 2648 |
+
"step": 377
|
| 2649 |
+
},
|
| 2650 |
+
{
|
| 2651 |
+
"epoch": 0.32390745501285345,
|
| 2652 |
+
"grad_norm": 0.015837060287594795,
|
| 2653 |
+
"learning_rate": 0.00015110477007916002,
|
| 2654 |
+
"loss": 0.1492,
|
| 2655 |
+
"step": 378
|
| 2656 |
+
},
|
| 2657 |
+
{
|
| 2658 |
+
"epoch": 0.32476435304198803,
|
| 2659 |
+
"grad_norm": 0.011251943185925484,
|
| 2660 |
+
"learning_rate": 0.0001487681669415742,
|
| 2661 |
+
"loss": 0.1505,
|
| 2662 |
+
"step": 379
|
| 2663 |
+
},
|
| 2664 |
+
{
|
| 2665 |
+
"epoch": 0.32562125107112255,
|
| 2666 |
+
"grad_norm": 0.011603351682424545,
|
| 2667 |
+
"learning_rate": 0.00014644660940672628,
|
| 2668 |
+
"loss": 0.1513,
|
| 2669 |
+
"step": 380
|
| 2670 |
+
},
|
| 2671 |
+
{
|
| 2672 |
+
"epoch": 0.3264781491002571,
|
| 2673 |
+
"grad_norm": 0.01359818410128355,
|
| 2674 |
+
"learning_rate": 0.00014414019692241437,
|
| 2675 |
+
"loss": 0.1459,
|
| 2676 |
+
"step": 381
|
| 2677 |
+
},
|
| 2678 |
+
{
|
| 2679 |
+
"epoch": 0.3273350471293916,
|
| 2680 |
+
"grad_norm": 0.012471762485802174,
|
| 2681 |
+
"learning_rate": 0.00014184902828767287,
|
| 2682 |
+
"loss": 0.1606,
|
| 2683 |
+
"step": 382
|
| 2684 |
+
},
|
| 2685 |
+
{
|
| 2686 |
+
"epoch": 0.3281919451585261,
|
| 2687 |
+
"grad_norm": 0.01537733431905508,
|
| 2688 |
+
"learning_rate": 0.0001395732016485406,
|
| 2689 |
+
"loss": 0.1446,
|
| 2690 |
+
"step": 383
|
| 2691 |
+
},
|
| 2692 |
+
{
|
| 2693 |
+
"epoch": 0.32904884318766064,
|
| 2694 |
+
"grad_norm": 0.01161841582506895,
|
| 2695 |
+
"learning_rate": 0.0001373128144938563,
|
| 2696 |
+
"loss": 0.1504,
|
| 2697 |
+
"step": 384
|
| 2698 |
+
},
|
| 2699 |
+
{
|
| 2700 |
+
"epoch": 0.3299057412167952,
|
| 2701 |
+
"grad_norm": 0.011465324088931084,
|
| 2702 |
+
"learning_rate": 0.00013506796365108232,
|
| 2703 |
+
"loss": 0.145,
|
| 2704 |
+
"step": 385
|
| 2705 |
+
},
|
| 2706 |
+
{
|
| 2707 |
+
"epoch": 0.33076263924592975,
|
| 2708 |
+
"grad_norm": 0.011288580484688282,
|
| 2709 |
+
"learning_rate": 0.00013283874528215734,
|
| 2710 |
+
"loss": 0.1494,
|
| 2711 |
+
"step": 386
|
| 2712 |
+
},
|
| 2713 |
+
{
|
| 2714 |
+
"epoch": 0.33161953727506427,
|
| 2715 |
+
"grad_norm": 0.012868880294263363,
|
| 2716 |
+
"learning_rate": 0.00013062525487937698,
|
| 2717 |
+
"loss": 0.1548,
|
| 2718 |
+
"step": 387
|
| 2719 |
+
},
|
| 2720 |
+
{
|
| 2721 |
+
"epoch": 0.3324764353041988,
|
| 2722 |
+
"grad_norm": 0.012357845902442932,
|
| 2723 |
+
"learning_rate": 0.00012842758726130281,
|
| 2724 |
+
"loss": 0.1504,
|
| 2725 |
+
"step": 388
|
| 2726 |
+
},
|
| 2727 |
+
{
|
| 2728 |
+
"epoch": 0.3333333333333333,
|
| 2729 |
+
"grad_norm": 0.011807232163846493,
|
| 2730 |
+
"learning_rate": 0.00012624583656870153,
|
| 2731 |
+
"loss": 0.1592,
|
| 2732 |
+
"step": 389
|
| 2733 |
+
},
|
| 2734 |
+
{
|
| 2735 |
+
"epoch": 0.3341902313624679,
|
| 2736 |
+
"grad_norm": 0.01140469592064619,
|
| 2737 |
+
"learning_rate": 0.00012408009626051135,
|
| 2738 |
+
"loss": 0.1577,
|
| 2739 |
+
"step": 390
|
| 2740 |
+
},
|
| 2741 |
+
{
|
| 2742 |
+
"epoch": 0.3350471293916024,
|
| 2743 |
+
"grad_norm": 0.009883826598525047,
|
| 2744 |
+
"learning_rate": 0.00012193045910983863,
|
| 2745 |
+
"loss": 0.1451,
|
| 2746 |
+
"step": 391
|
| 2747 |
+
},
|
| 2748 |
+
{
|
| 2749 |
+
"epoch": 0.33590402742073694,
|
| 2750 |
+
"grad_norm": 0.01367274671792984,
|
| 2751 |
+
"learning_rate": 0.00011979701719998454,
|
| 2752 |
+
"loss": 0.1574,
|
| 2753 |
+
"step": 392
|
| 2754 |
+
},
|
| 2755 |
+
{
|
| 2756 |
+
"epoch": 0.33676092544987146,
|
| 2757 |
+
"grad_norm": 0.011320062913000584,
|
| 2758 |
+
"learning_rate": 0.00011767986192049984,
|
| 2759 |
+
"loss": 0.1507,
|
| 2760 |
+
"step": 393
|
| 2761 |
+
},
|
| 2762 |
+
{
|
| 2763 |
+
"epoch": 0.337617823479006,
|
| 2764 |
+
"grad_norm": 0.011477210558950901,
|
| 2765 |
+
"learning_rate": 0.00011557908396327027,
|
| 2766 |
+
"loss": 0.1533,
|
| 2767 |
+
"step": 394
|
| 2768 |
+
},
|
| 2769 |
+
{
|
| 2770 |
+
"epoch": 0.3384747215081405,
|
| 2771 |
+
"grad_norm": 0.012028141878545284,
|
| 2772 |
+
"learning_rate": 0.00011349477331863151,
|
| 2773 |
+
"loss": 0.1592,
|
| 2774 |
+
"step": 395
|
| 2775 |
+
},
|
| 2776 |
+
{
|
| 2777 |
+
"epoch": 0.3393316195372751,
|
| 2778 |
+
"grad_norm": 0.01126360334455967,
|
| 2779 |
+
"learning_rate": 0.00011142701927151455,
|
| 2780 |
+
"loss": 0.1588,
|
| 2781 |
+
"step": 396
|
| 2782 |
+
},
|
| 2783 |
+
{
|
| 2784 |
+
"epoch": 0.3401885175664096,
|
| 2785 |
+
"grad_norm": 0.010625923052430153,
|
| 2786 |
+
"learning_rate": 0.00010937591039762085,
|
| 2787 |
+
"loss": 0.1443,
|
| 2788 |
+
"step": 397
|
| 2789 |
+
},
|
| 2790 |
+
{
|
| 2791 |
+
"epoch": 0.34104541559554413,
|
| 2792 |
+
"grad_norm": 0.01095986645668745,
|
| 2793 |
+
"learning_rate": 0.00010734153455962764,
|
| 2794 |
+
"loss": 0.1533,
|
| 2795 |
+
"step": 398
|
| 2796 |
+
},
|
| 2797 |
+
{
|
| 2798 |
+
"epoch": 0.34190231362467866,
|
| 2799 |
+
"grad_norm": 0.01167625468224287,
|
| 2800 |
+
"learning_rate": 0.00010532397890342504,
|
| 2801 |
+
"loss": 0.1465,
|
| 2802 |
+
"step": 399
|
| 2803 |
+
},
|
| 2804 |
+
{
|
| 2805 |
+
"epoch": 0.3427592116538132,
|
| 2806 |
+
"grad_norm": 0.010631192475557327,
|
| 2807 |
+
"learning_rate": 0.00010332332985438247,
|
| 2808 |
+
"loss": 0.1505,
|
| 2809 |
+
"step": 400
|
| 2810 |
+
},
|
| 2811 |
+
{
|
| 2812 |
+
"epoch": 0.34361610968294776,
|
| 2813 |
+
"grad_norm": 0.013422117568552494,
|
| 2814 |
+
"learning_rate": 0.0001013396731136465,
|
| 2815 |
+
"loss": 0.1475,
|
| 2816 |
+
"step": 401
|
| 2817 |
+
},
|
| 2818 |
+
{
|
| 2819 |
+
"epoch": 0.3444730077120823,
|
| 2820 |
+
"grad_norm": 0.015502565540373325,
|
| 2821 |
+
"learning_rate": 9.937309365446973e-05,
|
| 2822 |
+
"loss": 0.1564,
|
| 2823 |
+
"step": 402
|
| 2824 |
+
},
|
| 2825 |
+
{
|
| 2826 |
+
"epoch": 0.3453299057412168,
|
| 2827 |
+
"grad_norm": 0.012849084101617336,
|
| 2828 |
+
"learning_rate": 9.742367571857092e-05,
|
| 2829 |
+
"loss": 0.1492,
|
| 2830 |
+
"step": 403
|
| 2831 |
+
},
|
| 2832 |
+
{
|
| 2833 |
+
"epoch": 0.3461868037703513,
|
| 2834 |
+
"grad_norm": 0.011048342101275921,
|
| 2835 |
+
"learning_rate": 9.549150281252633e-05,
|
| 2836 |
+
"loss": 0.1497,
|
| 2837 |
+
"step": 404
|
| 2838 |
+
},
|
| 2839 |
+
{
|
| 2840 |
+
"epoch": 0.34704370179948585,
|
| 2841 |
+
"grad_norm": 0.00987666193395853,
|
| 2842 |
+
"learning_rate": 9.357665770419243e-05,
|
| 2843 |
+
"loss": 0.1449,
|
| 2844 |
+
"step": 405
|
| 2845 |
+
},
|
| 2846 |
+
{
|
| 2847 |
+
"epoch": 0.34790059982862037,
|
| 2848 |
+
"grad_norm": 0.009926311671733856,
|
| 2849 |
+
"learning_rate": 9.167922241916055e-05,
|
| 2850 |
+
"loss": 0.16,
|
| 2851 |
+
"step": 406
|
| 2852 |
+
},
|
| 2853 |
+
{
|
| 2854 |
+
"epoch": 0.34875749785775495,
|
| 2855 |
+
"grad_norm": 0.010267515666782856,
|
| 2856 |
+
"learning_rate": 8.979927823724321e-05,
|
| 2857 |
+
"loss": 0.1487,
|
| 2858 |
+
"step": 407
|
| 2859 |
+
},
|
| 2860 |
+
{
|
| 2861 |
+
"epoch": 0.3496143958868895,
|
| 2862 |
+
"grad_norm": 0.0110161192715168,
|
| 2863 |
+
"learning_rate": 8.793690568899215e-05,
|
| 2864 |
+
"loss": 0.1473,
|
| 2865 |
+
"step": 408
|
| 2866 |
+
},
|
| 2867 |
+
{
|
| 2868 |
+
"epoch": 0.350471293916024,
|
| 2869 |
+
"grad_norm": 0.009565568529069424,
|
| 2870 |
+
"learning_rate": 8.609218455224893e-05,
|
| 2871 |
+
"loss": 0.1462,
|
| 2872 |
+
"step": 409
|
| 2873 |
+
},
|
| 2874 |
+
{
|
| 2875 |
+
"epoch": 0.3513281919451585,
|
| 2876 |
+
"grad_norm": 0.00971953570842743,
|
| 2877 |
+
"learning_rate": 8.426519384872733e-05,
|
| 2878 |
+
"loss": 0.1435,
|
| 2879 |
+
"step": 410
|
| 2880 |
+
},
|
| 2881 |
+
{
|
| 2882 |
+
"epoch": 0.35218508997429304,
|
| 2883 |
+
"grad_norm": 0.00972969550639391,
|
| 2884 |
+
"learning_rate": 8.24560118406285e-05,
|
| 2885 |
+
"loss": 0.1482,
|
| 2886 |
+
"step": 411
|
| 2887 |
+
},
|
| 2888 |
+
{
|
| 2889 |
+
"epoch": 0.35304198800342756,
|
| 2890 |
+
"grad_norm": 0.012054841965436935,
|
| 2891 |
+
"learning_rate": 8.066471602728804e-05,
|
| 2892 |
+
"loss": 0.1463,
|
| 2893 |
+
"step": 412
|
| 2894 |
+
},
|
| 2895 |
+
{
|
| 2896 |
+
"epoch": 0.35389888603256214,
|
| 2897 |
+
"grad_norm": 0.011393013410270214,
|
| 2898 |
+
"learning_rate": 7.889138314185678e-05,
|
| 2899 |
+
"loss": 0.1485,
|
| 2900 |
+
"step": 413
|
| 2901 |
+
},
|
| 2902 |
+
{
|
| 2903 |
+
"epoch": 0.35475578406169667,
|
| 2904 |
+
"grad_norm": 0.011225726455450058,
|
| 2905 |
+
"learning_rate": 7.71360891480134e-05,
|
| 2906 |
+
"loss": 0.1452,
|
| 2907 |
+
"step": 414
|
| 2908 |
+
},
|
| 2909 |
+
{
|
| 2910 |
+
"epoch": 0.3556126820908312,
|
| 2911 |
+
"grad_norm": 0.01023333054035902,
|
| 2912 |
+
"learning_rate": 7.53989092367106e-05,
|
| 2913 |
+
"loss": 0.1358,
|
| 2914 |
+
"step": 415
|
| 2915 |
+
},
|
| 2916 |
+
{
|
| 2917 |
+
"epoch": 0.3564695801199657,
|
| 2918 |
+
"grad_norm": 0.010747382417321205,
|
| 2919 |
+
"learning_rate": 7.367991782295391e-05,
|
| 2920 |
+
"loss": 0.1427,
|
| 2921 |
+
"step": 416
|
| 2922 |
+
},
|
| 2923 |
+
{
|
| 2924 |
+
"epoch": 0.35732647814910024,
|
| 2925 |
+
"grad_norm": 0.01004976499825716,
|
| 2926 |
+
"learning_rate": 7.197918854261431e-05,
|
| 2927 |
+
"loss": 0.1473,
|
| 2928 |
+
"step": 417
|
| 2929 |
+
},
|
| 2930 |
+
{
|
| 2931 |
+
"epoch": 0.3581833761782348,
|
| 2932 |
+
"grad_norm": 0.009513266384601593,
|
| 2933 |
+
"learning_rate": 7.029679424927366e-05,
|
| 2934 |
+
"loss": 0.1493,
|
| 2935 |
+
"step": 418
|
| 2936 |
+
},
|
| 2937 |
+
{
|
| 2938 |
+
"epoch": 0.35904027420736934,
|
| 2939 |
+
"grad_norm": 0.015416925773024559,
|
| 2940 |
+
"learning_rate": 6.863280701110408e-05,
|
| 2941 |
+
"loss": 0.14,
|
| 2942 |
+
"step": 419
|
| 2943 |
+
},
|
| 2944 |
+
{
|
| 2945 |
+
"epoch": 0.35989717223650386,
|
| 2946 |
+
"grad_norm": 0.010445120744407177,
|
| 2947 |
+
"learning_rate": 6.698729810778065e-05,
|
| 2948 |
+
"loss": 0.1534,
|
| 2949 |
+
"step": 420
|
| 2950 |
+
},
|
| 2951 |
+
{
|
| 2952 |
+
"epoch": 0.3607540702656384,
|
| 2953 |
+
"grad_norm": 0.011475526727735996,
|
| 2954 |
+
"learning_rate": 6.536033802742814e-05,
|
| 2955 |
+
"loss": 0.1501,
|
| 2956 |
+
"step": 421
|
| 2957 |
+
},
|
| 2958 |
+
{
|
| 2959 |
+
"epoch": 0.3616109682947729,
|
| 2960 |
+
"grad_norm": 0.011237034574151039,
|
| 2961 |
+
"learning_rate": 6.375199646360142e-05,
|
| 2962 |
+
"loss": 0.1542,
|
| 2963 |
+
"step": 422
|
| 2964 |
+
},
|
| 2965 |
+
{
|
| 2966 |
+
"epoch": 0.36246786632390743,
|
| 2967 |
+
"grad_norm": 0.03411533311009407,
|
| 2968 |
+
"learning_rate": 6.21623423123001e-05,
|
| 2969 |
+
"loss": 0.2254,
|
| 2970 |
+
"step": 423
|
| 2971 |
+
},
|
| 2972 |
+
{
|
| 2973 |
+
"epoch": 0.363324764353042,
|
| 2974 |
+
"grad_norm": 0.009747683070600033,
|
| 2975 |
+
"learning_rate": 6.059144366901737e-05,
|
| 2976 |
+
"loss": 0.1454,
|
| 2977 |
+
"step": 424
|
| 2978 |
+
},
|
| 2979 |
+
{
|
| 2980 |
+
"epoch": 0.36418166238217653,
|
| 2981 |
+
"grad_norm": 0.010333815589547157,
|
| 2982 |
+
"learning_rate": 5.903936782582253e-05,
|
| 2983 |
+
"loss": 0.1479,
|
| 2984 |
+
"step": 425
|
| 2985 |
+
},
|
| 2986 |
+
{
|
| 2987 |
+
"epoch": 0.36503856041131105,
|
| 2988 |
+
"grad_norm": 0.010389740578830242,
|
| 2989 |
+
"learning_rate": 5.750618126847912e-05,
|
| 2990 |
+
"loss": 0.1504,
|
| 2991 |
+
"step": 426
|
| 2992 |
+
},
|
| 2993 |
+
{
|
| 2994 |
+
"epoch": 0.3658954584404456,
|
| 2995 |
+
"grad_norm": 0.010068557225167751,
|
| 2996 |
+
"learning_rate": 5.599194967359639e-05,
|
| 2997 |
+
"loss": 0.146,
|
| 2998 |
+
"step": 427
|
| 2999 |
+
},
|
| 3000 |
+
{
|
| 3001 |
+
"epoch": 0.3667523564695801,
|
| 3002 |
+
"grad_norm": 0.012611080892384052,
|
| 3003 |
+
"learning_rate": 5.449673790581611e-05,
|
| 3004 |
+
"loss": 0.1587,
|
| 3005 |
+
"step": 428
|
| 3006 |
+
},
|
| 3007 |
+
{
|
| 3008 |
+
"epoch": 0.3676092544987147,
|
| 3009 |
+
"grad_norm": 0.010398144833743572,
|
| 3010 |
+
"learning_rate": 5.3020610015033946e-05,
|
| 3011 |
+
"loss": 0.1455,
|
| 3012 |
+
"step": 429
|
| 3013 |
+
},
|
| 3014 |
+
{
|
| 3015 |
+
"epoch": 0.3684661525278492,
|
| 3016 |
+
"grad_norm": 0.012510698288679123,
|
| 3017 |
+
"learning_rate": 5.1563629233655876e-05,
|
| 3018 |
+
"loss": 0.1544,
|
| 3019 |
+
"step": 430
|
| 3020 |
+
},
|
| 3021 |
+
{
|
| 3022 |
+
"epoch": 0.3693230505569837,
|
| 3023 |
+
"grad_norm": 0.009702283889055252,
|
| 3024 |
+
"learning_rate": 5.0125857973889355e-05,
|
| 3025 |
+
"loss": 0.1376,
|
| 3026 |
+
"step": 431
|
| 3027 |
+
},
|
| 3028 |
+
{
|
| 3029 |
+
"epoch": 0.37017994858611825,
|
| 3030 |
+
"grad_norm": 0.015176232904195786,
|
| 3031 |
+
"learning_rate": 4.87073578250698e-05,
|
| 3032 |
+
"loss": 0.1505,
|
| 3033 |
+
"step": 432
|
| 3034 |
+
},
|
| 3035 |
+
{
|
| 3036 |
+
"epoch": 0.37103684661525277,
|
| 3037 |
+
"grad_norm": 0.009702375158667564,
|
| 3038 |
+
"learning_rate": 4.730818955102234e-05,
|
| 3039 |
+
"loss": 0.1465,
|
| 3040 |
+
"step": 433
|
| 3041 |
+
},
|
| 3042 |
+
{
|
| 3043 |
+
"epoch": 0.3718937446443873,
|
| 3044 |
+
"grad_norm": 0.01697813719511032,
|
| 3045 |
+
"learning_rate": 4.592841308745932e-05,
|
| 3046 |
+
"loss": 0.1528,
|
| 3047 |
+
"step": 434
|
| 3048 |
+
},
|
| 3049 |
+
{
|
| 3050 |
+
"epoch": 0.37275064267352187,
|
| 3051 |
+
"grad_norm": 0.01560263428837061,
|
| 3052 |
+
"learning_rate": 4.456808753941205e-05,
|
| 3053 |
+
"loss": 0.1437,
|
| 3054 |
+
"step": 435
|
| 3055 |
+
},
|
| 3056 |
+
{
|
| 3057 |
+
"epoch": 0.3736075407026564,
|
| 3058 |
+
"grad_norm": 0.00954125914722681,
|
| 3059 |
+
"learning_rate": 4.322727117869951e-05,
|
| 3060 |
+
"loss": 0.1501,
|
| 3061 |
+
"step": 436
|
| 3062 |
+
},
|
| 3063 |
+
{
|
| 3064 |
+
"epoch": 0.3744644387317909,
|
| 3065 |
+
"grad_norm": 0.01739114709198475,
|
| 3066 |
+
"learning_rate": 4.190602144143207e-05,
|
| 3067 |
+
"loss": 0.1568,
|
| 3068 |
+
"step": 437
|
| 3069 |
+
},
|
| 3070 |
+
{
|
| 3071 |
+
"epoch": 0.37532133676092544,
|
| 3072 |
+
"grad_norm": 0.011187477968633175,
|
| 3073 |
+
"learning_rate": 4.06043949255509e-05,
|
| 3074 |
+
"loss": 0.1558,
|
| 3075 |
+
"step": 438
|
| 3076 |
+
},
|
| 3077 |
+
{
|
| 3078 |
+
"epoch": 0.37617823479005996,
|
| 3079 |
+
"grad_norm": 0.01066075824201107,
|
| 3080 |
+
"learning_rate": 3.932244738840379e-05,
|
| 3081 |
+
"loss": 0.1458,
|
| 3082 |
+
"step": 439
|
| 3083 |
+
},
|
| 3084 |
+
{
|
| 3085 |
+
"epoch": 0.37703513281919454,
|
| 3086 |
+
"grad_norm": 0.009765625,
|
| 3087 |
+
"learning_rate": 3.806023374435663e-05,
|
| 3088 |
+
"loss": 0.1492,
|
| 3089 |
+
"step": 440
|
| 3090 |
+
},
|
| 3091 |
+
{
|
| 3092 |
+
"epoch": 0.37789203084832906,
|
| 3093 |
+
"grad_norm": 0.009504380635917187,
|
| 3094 |
+
"learning_rate": 3.681780806244095e-05,
|
| 3095 |
+
"loss": 0.1413,
|
| 3096 |
+
"step": 441
|
| 3097 |
+
},
|
| 3098 |
+
{
|
| 3099 |
+
"epoch": 0.3787489288774636,
|
| 3100 |
+
"grad_norm": 0.010201869532465935,
|
| 3101 |
+
"learning_rate": 3.559522356403788e-05,
|
| 3102 |
+
"loss": 0.1486,
|
| 3103 |
+
"step": 442
|
| 3104 |
+
},
|
| 3105 |
+
{
|
| 3106 |
+
"epoch": 0.3796058269065981,
|
| 3107 |
+
"grad_norm": 0.010059897787868977,
|
| 3108 |
+
"learning_rate": 3.439253262059822e-05,
|
| 3109 |
+
"loss": 0.1314,
|
| 3110 |
+
"step": 443
|
| 3111 |
+
},
|
| 3112 |
+
{
|
| 3113 |
+
"epoch": 0.38046272493573263,
|
| 3114 |
+
"grad_norm": 0.011777847073972225,
|
| 3115 |
+
"learning_rate": 3.3209786751399184e-05,
|
| 3116 |
+
"loss": 0.1499,
|
| 3117 |
+
"step": 444
|
| 3118 |
+
},
|
| 3119 |
+
{
|
| 3120 |
+
"epoch": 0.38131962296486716,
|
| 3121 |
+
"grad_norm": 0.00938204862177372,
|
| 3122 |
+
"learning_rate": 3.2047036621337236e-05,
|
| 3123 |
+
"loss": 0.1453,
|
| 3124 |
+
"step": 445
|
| 3125 |
+
},
|
| 3126 |
+
{
|
| 3127 |
+
"epoch": 0.38217652099400173,
|
| 3128 |
+
"grad_norm": 0.009476981125772,
|
| 3129 |
+
"learning_rate": 3.0904332038757974e-05,
|
| 3130 |
+
"loss": 0.1446,
|
| 3131 |
+
"step": 446
|
| 3132 |
+
},
|
| 3133 |
+
{
|
| 3134 |
+
"epoch": 0.38303341902313626,
|
| 3135 |
+
"grad_norm": 0.009422726929187775,
|
| 3136 |
+
"learning_rate": 2.9781721953322627e-05,
|
| 3137 |
+
"loss": 0.1488,
|
| 3138 |
+
"step": 447
|
| 3139 |
+
},
|
| 3140 |
+
{
|
| 3141 |
+
"epoch": 0.3838903170522708,
|
| 3142 |
+
"grad_norm": 0.00957945454865694,
|
| 3143 |
+
"learning_rate": 2.8679254453910786e-05,
|
| 3144 |
+
"loss": 0.1458,
|
| 3145 |
+
"step": 448
|
| 3146 |
+
},
|
| 3147 |
+
{
|
| 3148 |
+
"epoch": 0.3847472150814053,
|
| 3149 |
+
"grad_norm": 0.009238988161087036,
|
| 3150 |
+
"learning_rate": 2.7596976766560976e-05,
|
| 3151 |
+
"loss": 0.1468,
|
| 3152 |
+
"step": 449
|
| 3153 |
+
},
|
| 3154 |
+
{
|
| 3155 |
+
"epoch": 0.3856041131105398,
|
| 3156 |
+
"grad_norm": 0.010409279726445675,
|
| 3157 |
+
"learning_rate": 2.653493525244721e-05,
|
| 3158 |
+
"loss": 0.1488,
|
| 3159 |
+
"step": 450
|
| 3160 |
+
},
|
| 3161 |
+
{
|
| 3162 |
+
"epoch": 0.3864610111396744,
|
| 3163 |
+
"grad_norm": 0.010871903970837593,
|
| 3164 |
+
"learning_rate": 2.5493175405893076e-05,
|
| 3165 |
+
"loss": 0.1448,
|
| 3166 |
+
"step": 451
|
| 3167 |
+
},
|
| 3168 |
+
{
|
| 3169 |
+
"epoch": 0.3873179091688089,
|
| 3170 |
+
"grad_norm": 0.01018882729113102,
|
| 3171 |
+
"learning_rate": 2.4471741852423235e-05,
|
| 3172 |
+
"loss": 0.149,
|
| 3173 |
+
"step": 452
|
| 3174 |
+
},
|
| 3175 |
+
{
|
| 3176 |
+
"epoch": 0.38817480719794345,
|
| 3177 |
+
"grad_norm": 0.00999706145375967,
|
| 3178 |
+
"learning_rate": 2.3470678346851513e-05,
|
| 3179 |
+
"loss": 0.1476,
|
| 3180 |
+
"step": 453
|
| 3181 |
+
},
|
| 3182 |
+
{
|
| 3183 |
+
"epoch": 0.389031705227078,
|
| 3184 |
+
"grad_norm": 0.008825534954667091,
|
| 3185 |
+
"learning_rate": 2.2490027771406685e-05,
|
| 3186 |
+
"loss": 0.1522,
|
| 3187 |
+
"step": 454
|
| 3188 |
+
},
|
| 3189 |
+
{
|
| 3190 |
+
"epoch": 0.3898886032562125,
|
| 3191 |
+
"grad_norm": 0.010562725365161896,
|
| 3192 |
+
"learning_rate": 2.152983213389559e-05,
|
| 3193 |
+
"loss": 0.1477,
|
| 3194 |
+
"step": 455
|
| 3195 |
+
},
|
| 3196 |
+
{
|
| 3197 |
+
"epoch": 0.390745501285347,
|
| 3198 |
+
"grad_norm": 0.01029855664819479,
|
| 3199 |
+
"learning_rate": 2.0590132565903473e-05,
|
| 3200 |
+
"loss": 0.1603,
|
| 3201 |
+
"step": 456
|
| 3202 |
+
},
|
| 3203 |
+
{
|
| 3204 |
+
"epoch": 0.3916023993144816,
|
| 3205 |
+
"grad_norm": 0.009447697550058365,
|
| 3206 |
+
"learning_rate": 1.9670969321032406e-05,
|
| 3207 |
+
"loss": 0.1507,
|
| 3208 |
+
"step": 457
|
| 3209 |
+
},
|
| 3210 |
+
{
|
| 3211 |
+
"epoch": 0.3924592973436161,
|
| 3212 |
+
"grad_norm": 0.009440843015909195,
|
| 3213 |
+
"learning_rate": 1.8772381773176416e-05,
|
| 3214 |
+
"loss": 0.1487,
|
| 3215 |
+
"step": 458
|
| 3216 |
+
},
|
| 3217 |
+
{
|
| 3218 |
+
"epoch": 0.39331619537275064,
|
| 3219 |
+
"grad_norm": 0.01008307933807373,
|
| 3220 |
+
"learning_rate": 1.7894408414835363e-05,
|
| 3221 |
+
"loss": 0.1482,
|
| 3222 |
+
"step": 459
|
| 3223 |
+
},
|
| 3224 |
+
{
|
| 3225 |
+
"epoch": 0.39417309340188517,
|
| 3226 |
+
"grad_norm": 0.010164221748709679,
|
| 3227 |
+
"learning_rate": 1.70370868554659e-05,
|
| 3228 |
+
"loss": 0.1514,
|
| 3229 |
+
"step": 460
|
| 3230 |
+
},
|
| 3231 |
+
{
|
| 3232 |
+
"epoch": 0.3950299914310197,
|
| 3233 |
+
"grad_norm": 0.009882348589599133,
|
| 3234 |
+
"learning_rate": 1.620045381987012e-05,
|
| 3235 |
+
"loss": 0.1459,
|
| 3236 |
+
"step": 461
|
| 3237 |
+
},
|
| 3238 |
+
{
|
| 3239 |
+
"epoch": 0.39588688946015427,
|
| 3240 |
+
"grad_norm": 0.01052442193031311,
|
| 3241 |
+
"learning_rate": 1.538454514662285e-05,
|
| 3242 |
+
"loss": 0.1565,
|
| 3243 |
+
"step": 462
|
| 3244 |
+
},
|
| 3245 |
+
{
|
| 3246 |
+
"epoch": 0.3967437874892888,
|
| 3247 |
+
"grad_norm": 0.010268788784742355,
|
| 3248 |
+
"learning_rate": 1.4589395786535953e-05,
|
| 3249 |
+
"loss": 0.1491,
|
| 3250 |
+
"step": 463
|
| 3251 |
+
},
|
| 3252 |
+
{
|
| 3253 |
+
"epoch": 0.3976006855184233,
|
| 3254 |
+
"grad_norm": 0.00896680261939764,
|
| 3255 |
+
"learning_rate": 1.3815039801161721e-05,
|
| 3256 |
+
"loss": 0.1434,
|
| 3257 |
+
"step": 464
|
| 3258 |
+
},
|
| 3259 |
+
{
|
| 3260 |
+
"epoch": 0.39845758354755784,
|
| 3261 |
+
"grad_norm": 0.009579429402947426,
|
| 3262 |
+
"learning_rate": 1.3061510361333184e-05,
|
| 3263 |
+
"loss": 0.147,
|
| 3264 |
+
"step": 465
|
| 3265 |
+
},
|
| 3266 |
+
{
|
| 3267 |
+
"epoch": 0.39931448157669236,
|
| 3268 |
+
"grad_norm": 0.009059751406311989,
|
| 3269 |
+
"learning_rate": 1.232883974574367e-05,
|
| 3270 |
+
"loss": 0.1454,
|
| 3271 |
+
"step": 466
|
| 3272 |
+
},
|
| 3273 |
+
{
|
| 3274 |
+
"epoch": 0.4001713796058269,
|
| 3275 |
+
"grad_norm": 0.00879402831196785,
|
| 3276 |
+
"learning_rate": 1.1617059339563806e-05,
|
| 3277 |
+
"loss": 0.1465,
|
| 3278 |
+
"step": 467
|
| 3279 |
+
},
|
| 3280 |
+
{
|
| 3281 |
+
"epoch": 0.40102827763496146,
|
| 3282 |
+
"grad_norm": 0.008100686594843864,
|
| 3283 |
+
"learning_rate": 1.0926199633097156e-05,
|
| 3284 |
+
"loss": 0.1453,
|
| 3285 |
+
"step": 468
|
| 3286 |
+
},
|
| 3287 |
+
{
|
| 3288 |
+
"epoch": 0.401885175664096,
|
| 3289 |
+
"grad_norm": 0.011180667206645012,
|
| 3290 |
+
"learning_rate": 1.0256290220474307e-05,
|
| 3291 |
+
"loss": 0.155,
|
| 3292 |
+
"step": 469
|
| 3293 |
+
},
|
| 3294 |
+
{
|
| 3295 |
+
"epoch": 0.4027420736932305,
|
| 3296 |
+
"grad_norm": 0.009214532561600208,
|
| 3297 |
+
"learning_rate": 9.607359798384786e-06,
|
| 3298 |
+
"loss": 0.152,
|
| 3299 |
+
"step": 470
|
| 3300 |
+
},
|
| 3301 |
+
{
|
| 3302 |
+
"epoch": 0.40359897172236503,
|
| 3303 |
+
"grad_norm": 0.011793004348874092,
|
| 3304 |
+
"learning_rate": 8.979436164848088e-06,
|
| 3305 |
+
"loss": 0.1503,
|
| 3306 |
+
"step": 471
|
| 3307 |
+
},
|
| 3308 |
+
{
|
| 3309 |
+
"epoch": 0.40445586975149955,
|
| 3310 |
+
"grad_norm": 0.008904839865863323,
|
| 3311 |
+
"learning_rate": 8.372546218022748e-06,
|
| 3312 |
+
"loss": 0.1372,
|
| 3313 |
+
"step": 472
|
| 3314 |
+
},
|
| 3315 |
+
{
|
| 3316 |
+
"epoch": 0.40531276778063413,
|
| 3317 |
+
"grad_norm": 0.010819431394338608,
|
| 3318 |
+
"learning_rate": 7.786715955054202e-06,
|
| 3319 |
+
"loss": 0.1517,
|
| 3320 |
+
"step": 473
|
| 3321 |
+
},
|
| 3322 |
+
{
|
| 3323 |
+
"epoch": 0.40616966580976865,
|
| 3324 |
+
"grad_norm": 0.008758115582168102,
|
| 3325 |
+
"learning_rate": 7.221970470961125e-06,
|
| 3326 |
+
"loss": 0.1415,
|
| 3327 |
+
"step": 474
|
| 3328 |
+
},
|
| 3329 |
+
{
|
| 3330 |
+
"epoch": 0.4070265638389032,
|
| 3331 |
+
"grad_norm": 0.009194256737828255,
|
| 3332 |
+
"learning_rate": 6.678333957560512e-06,
|
| 3333 |
+
"loss": 0.1497,
|
| 3334 |
+
"step": 475
|
| 3335 |
+
},
|
| 3336 |
+
{
|
| 3337 |
+
"epoch": 0.4078834618680377,
|
| 3338 |
+
"grad_norm": 0.010541984811425209,
|
| 3339 |
+
"learning_rate": 6.15582970243117e-06,
|
| 3340 |
+
"loss": 0.155,
|
| 3341 |
+
"step": 476
|
| 3342 |
+
},
|
| 3343 |
+
{
|
| 3344 |
+
"epoch": 0.4087403598971722,
|
| 3345 |
+
"grad_norm": 0.009917198680341244,
|
| 3346 |
+
"learning_rate": 5.6544800879163026e-06,
|
| 3347 |
+
"loss": 0.1483,
|
| 3348 |
+
"step": 477
|
| 3349 |
+
},
|
| 3350 |
+
{
|
| 3351 |
+
"epoch": 0.40959725792630675,
|
| 3352 |
+
"grad_norm": 0.01124248094856739,
|
| 3353 |
+
"learning_rate": 5.174306590164879e-06,
|
| 3354 |
+
"loss": 0.1506,
|
| 3355 |
+
"step": 478
|
| 3356 |
+
},
|
| 3357 |
+
{
|
| 3358 |
+
"epoch": 0.4104541559554413,
|
| 3359 |
+
"grad_norm": 0.009757012128829956,
|
| 3360 |
+
"learning_rate": 4.715329778211374e-06,
|
| 3361 |
+
"loss": 0.1448,
|
| 3362 |
+
"step": 479
|
| 3363 |
+
},
|
| 3364 |
+
{
|
| 3365 |
+
"epoch": 0.41131105398457585,
|
| 3366 |
+
"grad_norm": 0.009212766773998737,
|
| 3367 |
+
"learning_rate": 4.277569313094809e-06,
|
| 3368 |
+
"loss": 0.1483,
|
| 3369 |
+
"step": 480
|
| 3370 |
+
},
|
| 3371 |
+
{
|
| 3372 |
+
"epoch": 0.41216795201371037,
|
| 3373 |
+
"grad_norm": 0.01022613886743784,
|
| 3374 |
+
"learning_rate": 3.861043947016474e-06,
|
| 3375 |
+
"loss": 0.1514,
|
| 3376 |
+
"step": 481
|
| 3377 |
+
},
|
| 3378 |
+
{
|
| 3379 |
+
"epoch": 0.4130248500428449,
|
| 3380 |
+
"grad_norm": 0.012525631114840508,
|
| 3381 |
+
"learning_rate": 3.4657715225368535e-06,
|
| 3382 |
+
"loss": 0.1523,
|
| 3383 |
+
"step": 482
|
| 3384 |
+
},
|
| 3385 |
+
{
|
| 3386 |
+
"epoch": 0.4138817480719794,
|
| 3387 |
+
"grad_norm": 0.013477517291903496,
|
| 3388 |
+
"learning_rate": 3.09176897181096e-06,
|
| 3389 |
+
"loss": 0.1508,
|
| 3390 |
+
"step": 483
|
| 3391 |
+
},
|
| 3392 |
+
{
|
| 3393 |
+
"epoch": 0.414738646101114,
|
| 3394 |
+
"grad_norm": 0.009571454487740993,
|
| 3395 |
+
"learning_rate": 2.739052315863355e-06,
|
| 3396 |
+
"loss": 0.1472,
|
| 3397 |
+
"step": 484
|
| 3398 |
+
},
|
| 3399 |
+
{
|
| 3400 |
+
"epoch": 0.4155955441302485,
|
| 3401 |
+
"grad_norm": 0.008747117593884468,
|
| 3402 |
+
"learning_rate": 2.4076366639015913e-06,
|
| 3403 |
+
"loss": 0.149,
|
| 3404 |
+
"step": 485
|
| 3405 |
+
},
|
| 3406 |
+
{
|
| 3407 |
+
"epoch": 0.41645244215938304,
|
| 3408 |
+
"grad_norm": 0.008356280624866486,
|
| 3409 |
+
"learning_rate": 2.097536212669171e-06,
|
| 3410 |
+
"loss": 0.1475,
|
| 3411 |
+
"step": 486
|
| 3412 |
+
},
|
| 3413 |
+
{
|
| 3414 |
+
"epoch": 0.41730934018851756,
|
| 3415 |
+
"grad_norm": 0.008765033446252346,
|
| 3416 |
+
"learning_rate": 1.8087642458373132e-06,
|
| 3417 |
+
"loss": 0.1476,
|
| 3418 |
+
"step": 487
|
| 3419 |
+
},
|
| 3420 |
+
{
|
| 3421 |
+
"epoch": 0.4181662382176521,
|
| 3422 |
+
"grad_norm": 0.009499343112111092,
|
| 3423 |
+
"learning_rate": 1.541333133436018e-06,
|
| 3424 |
+
"loss": 0.1429,
|
| 3425 |
+
"step": 488
|
| 3426 |
+
},
|
| 3427 |
+
{
|
| 3428 |
+
"epoch": 0.4190231362467866,
|
| 3429 |
+
"grad_norm": 0.009461235255002975,
|
| 3430 |
+
"learning_rate": 1.2952543313240472e-06,
|
| 3431 |
+
"loss": 0.1467,
|
| 3432 |
+
"step": 489
|
| 3433 |
+
},
|
| 3434 |
+
{
|
| 3435 |
+
"epoch": 0.4198800342759212,
|
| 3436 |
+
"grad_norm": 0.008559729903936386,
|
| 3437 |
+
"learning_rate": 1.0705383806982606e-06,
|
| 3438 |
+
"loss": 0.1529,
|
| 3439 |
+
"step": 490
|
| 3440 |
+
},
|
| 3441 |
+
{
|
| 3442 |
+
"epoch": 0.4207369323050557,
|
| 3443 |
+
"grad_norm": 0.010164659470319748,
|
| 3444 |
+
"learning_rate": 8.671949076420882e-07,
|
| 3445 |
+
"loss": 0.1567,
|
| 3446 |
+
"step": 491
|
| 3447 |
+
},
|
| 3448 |
+
{
|
| 3449 |
+
"epoch": 0.42159383033419023,
|
| 3450 |
+
"grad_norm": 0.01175006665289402,
|
| 3451 |
+
"learning_rate": 6.852326227130834e-07,
|
| 3452 |
+
"loss": 0.1579,
|
| 3453 |
+
"step": 492
|
| 3454 |
+
},
|
| 3455 |
+
{
|
| 3456 |
+
"epoch": 0.42245072836332476,
|
| 3457 |
+
"grad_norm": 0.013031876645982265,
|
| 3458 |
+
"learning_rate": 5.246593205699424e-07,
|
| 3459 |
+
"loss": 0.1423,
|
| 3460 |
+
"step": 493
|
| 3461 |
+
},
|
| 3462 |
+
{
|
| 3463 |
+
"epoch": 0.4233076263924593,
|
| 3464 |
+
"grad_norm": 0.009586134925484657,
|
| 3465 |
+
"learning_rate": 3.854818796385495e-07,
|
| 3466 |
+
"loss": 0.1539,
|
| 3467 |
+
"step": 494
|
| 3468 |
+
},
|
| 3469 |
+
{
|
| 3470 |
+
"epoch": 0.4241645244215938,
|
| 3471 |
+
"grad_norm": 0.009407884441316128,
|
| 3472 |
+
"learning_rate": 2.677062618171577e-07,
|
| 3473 |
+
"loss": 0.1446,
|
| 3474 |
+
"step": 495
|
| 3475 |
+
},
|
| 3476 |
+
{
|
| 3477 |
+
"epoch": 0.4250214224507284,
|
| 3478 |
+
"grad_norm": 0.00942971371114254,
|
| 3479 |
+
"learning_rate": 1.7133751222137007e-07,
|
| 3480 |
+
"loss": 0.1533,
|
| 3481 |
+
"step": 496
|
| 3482 |
+
},
|
| 3483 |
+
{
|
| 3484 |
+
"epoch": 0.4258783204798629,
|
| 3485 |
+
"grad_norm": 0.009059898555278778,
|
| 3486 |
+
"learning_rate": 9.637975896759077e-08,
|
| 3487 |
+
"loss": 0.1403,
|
| 3488 |
+
"step": 497
|
| 3489 |
+
},
|
| 3490 |
+
{
|
| 3491 |
+
"epoch": 0.4267352185089974,
|
| 3492 |
+
"grad_norm": 0.010631782002747059,
|
| 3493 |
+
"learning_rate": 4.283621299649987e-08,
|
| 3494 |
+
"loss": 0.159,
|
| 3495 |
+
"step": 498
|
| 3496 |
+
},
|
| 3497 |
+
{
|
| 3498 |
+
"epoch": 0.42759211653813195,
|
| 3499 |
+
"grad_norm": 0.009222394786775112,
|
| 3500 |
+
"learning_rate": 1.0709167935385456e-08,
|
| 3501 |
+
"loss": 0.1512,
|
| 3502 |
+
"step": 499
|
| 3503 |
+
},
|
| 3504 |
+
{
|
| 3505 |
+
"epoch": 0.4284490145672665,
|
| 3506 |
+
"grad_norm": 0.008665120229125023,
|
| 3507 |
+
"learning_rate": 0.0,
|
| 3508 |
+
"loss": 0.1478,
|
| 3509 |
+
"step": 500
|
| 3510 |
+
},
|
| 3511 |
+
{
|
| 3512 |
+
"epoch": 0.4284490145672665,
|
| 3513 |
+
"step": 500,
|
| 3514 |
+
"total_flos": 1.0100648106681958e+20,
|
| 3515 |
+
"train_loss": 0.25743724408745766,
|
| 3516 |
+
"train_runtime": 32795.7548,
|
| 3517 |
+
"train_samples_per_second": 0.244,
|
| 3518 |
+
"train_steps_per_second": 0.015
|
| 3519 |
+
}
|
| 3520 |
+
],
|
| 3521 |
+
"logging_steps": 1.0,
|
| 3522 |
+
"max_steps": 500,
|
| 3523 |
+
"num_input_tokens_seen": 0,
|
| 3524 |
+
"num_train_epochs": 1,
|
| 3525 |
+
"save_steps": 500,
|
| 3526 |
+
"stateful_callbacks": {
|
| 3527 |
+
"TrainerControl": {
|
| 3528 |
+
"args": {
|
| 3529 |
+
"should_epoch_stop": false,
|
| 3530 |
+
"should_evaluate": false,
|
| 3531 |
+
"should_log": false,
|
| 3532 |
+
"should_save": false,
|
| 3533 |
+
"should_training_stop": false
|
| 3534 |
+
},
|
| 3535 |
+
"attributes": {}
|
| 3536 |
+
}
|
| 3537 |
+
},
|
| 3538 |
+
"total_flos": 1.0100648106681958e+20,
|
| 3539 |
+
"train_batch_size": 1,
|
| 3540 |
+
"trial_name": null,
|
| 3541 |
+
"trial_params": null
|
| 3542 |
+
}
|