Training in progress, step 2036, checkpoint
Browse files- last-checkpoint/config.json +28 -0
- last-checkpoint/generation_config.json +12 -0
- last-checkpoint/model.safetensors +3 -0
- last-checkpoint/optimizer.pt +3 -0
- last-checkpoint/rng_state.pth +3 -0
- last-checkpoint/scheduler.pt +3 -0
- last-checkpoint/trainer_state.json +1455 -0
- last-checkpoint/training_args.bin +3 -0
last-checkpoint/config.json
ADDED
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"architectures": [
|
3 |
+
"Qwen2ForCausalLM"
|
4 |
+
],
|
5 |
+
"attention_dropout": 0.0,
|
6 |
+
"bos_token_id": 151643,
|
7 |
+
"eos_token_id": 151645,
|
8 |
+
"hidden_act": "silu",
|
9 |
+
"hidden_size": 2048,
|
10 |
+
"initializer_range": 0.02,
|
11 |
+
"intermediate_size": 5504,
|
12 |
+
"max_position_embeddings": 32768,
|
13 |
+
"max_window_layers": 21,
|
14 |
+
"model_type": "qwen2",
|
15 |
+
"num_attention_heads": 16,
|
16 |
+
"num_hidden_layers": 24,
|
17 |
+
"num_key_value_heads": 16,
|
18 |
+
"rms_norm_eps": 1e-06,
|
19 |
+
"rope_scaling": null,
|
20 |
+
"rope_theta": 1000000.0,
|
21 |
+
"sliding_window": 32768,
|
22 |
+
"tie_word_embeddings": false,
|
23 |
+
"torch_dtype": "bfloat16",
|
24 |
+
"transformers_version": "4.51.3",
|
25 |
+
"use_cache": true,
|
26 |
+
"use_sliding_window": false,
|
27 |
+
"vocab_size": 151936
|
28 |
+
}
|
last-checkpoint/generation_config.json
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token_id": 151643,
|
3 |
+
"do_sample": true,
|
4 |
+
"eos_token_id": [
|
5 |
+
151645,
|
6 |
+
151643
|
7 |
+
],
|
8 |
+
"pad_token_id": 151643,
|
9 |
+
"repetition_penalty": 1.1,
|
10 |
+
"top_p": 0.8,
|
11 |
+
"transformers_version": "4.51.3"
|
12 |
+
}
|
last-checkpoint/model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e02ee0e3091a12bc5abb031e386a74a22cb9bdc622a9842d1c9722a565b8c7c1
|
3 |
+
size 3673690696
|
last-checkpoint/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:eb86ff67aa4c157e4b6eff3a0004aa92b27aa127bec34b95e5c548603ce0a291
|
3 |
+
size 7347565686
|
last-checkpoint/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:9d9cd6a0487226e5bd30d1846894c82af483733ab4381b75bae9c0745e05d405
|
3 |
+
size 14244
|
last-checkpoint/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:09933cc0b85c7ce3d86d14a8c13caccc68575168a9217284ecd221f1cc79411a
|
3 |
+
size 1064
|
last-checkpoint/trainer_state.json
ADDED
@@ -0,0 +1,1455 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_global_step": null,
|
3 |
+
"best_metric": null,
|
4 |
+
"best_model_checkpoint": null,
|
5 |
+
"epoch": 1.0,
|
6 |
+
"eval_steps": 500,
|
7 |
+
"global_step": 2036,
|
8 |
+
"is_hyper_param_search": false,
|
9 |
+
"is_local_process_zero": true,
|
10 |
+
"is_world_process_zero": true,
|
11 |
+
"log_history": [
|
12 |
+
{
|
13 |
+
"epoch": 0.004911591355599214,
|
14 |
+
"grad_norm": 4.0625,
|
15 |
+
"learning_rate": 4.977897838899804e-05,
|
16 |
+
"loss": 2.9953,
|
17 |
+
"step": 10
|
18 |
+
},
|
19 |
+
{
|
20 |
+
"epoch": 0.009823182711198428,
|
21 |
+
"grad_norm": 1.6328125,
|
22 |
+
"learning_rate": 4.953339882121808e-05,
|
23 |
+
"loss": 0.0935,
|
24 |
+
"step": 20
|
25 |
+
},
|
26 |
+
{
|
27 |
+
"epoch": 0.014734774066797643,
|
28 |
+
"grad_norm": 1.578125,
|
29 |
+
"learning_rate": 4.928781925343812e-05,
|
30 |
+
"loss": 0.073,
|
31 |
+
"step": 30
|
32 |
+
},
|
33 |
+
{
|
34 |
+
"epoch": 0.019646365422396856,
|
35 |
+
"grad_norm": 1.2890625,
|
36 |
+
"learning_rate": 4.904223968565815e-05,
|
37 |
+
"loss": 0.0679,
|
38 |
+
"step": 40
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.02455795677799607,
|
42 |
+
"grad_norm": 1.5078125,
|
43 |
+
"learning_rate": 4.87966601178782e-05,
|
44 |
+
"loss": 0.055,
|
45 |
+
"step": 50
|
46 |
+
},
|
47 |
+
{
|
48 |
+
"epoch": 0.029469548133595286,
|
49 |
+
"grad_norm": 1.234375,
|
50 |
+
"learning_rate": 4.855108055009823e-05,
|
51 |
+
"loss": 0.0542,
|
52 |
+
"step": 60
|
53 |
+
},
|
54 |
+
{
|
55 |
+
"epoch": 0.0343811394891945,
|
56 |
+
"grad_norm": 0.984375,
|
57 |
+
"learning_rate": 4.830550098231827e-05,
|
58 |
+
"loss": 0.0458,
|
59 |
+
"step": 70
|
60 |
+
},
|
61 |
+
{
|
62 |
+
"epoch": 0.03929273084479371,
|
63 |
+
"grad_norm": 0.95703125,
|
64 |
+
"learning_rate": 4.805992141453831e-05,
|
65 |
+
"loss": 0.0485,
|
66 |
+
"step": 80
|
67 |
+
},
|
68 |
+
{
|
69 |
+
"epoch": 0.04420432220039293,
|
70 |
+
"grad_norm": 1.21875,
|
71 |
+
"learning_rate": 4.781434184675835e-05,
|
72 |
+
"loss": 0.0563,
|
73 |
+
"step": 90
|
74 |
+
},
|
75 |
+
{
|
76 |
+
"epoch": 0.04911591355599214,
|
77 |
+
"grad_norm": 0.85546875,
|
78 |
+
"learning_rate": 4.756876227897839e-05,
|
79 |
+
"loss": 0.0525,
|
80 |
+
"step": 100
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 0.054027504911591355,
|
84 |
+
"grad_norm": 1.09375,
|
85 |
+
"learning_rate": 4.732318271119843e-05,
|
86 |
+
"loss": 0.0501,
|
87 |
+
"step": 110
|
88 |
+
},
|
89 |
+
{
|
90 |
+
"epoch": 0.05893909626719057,
|
91 |
+
"grad_norm": 0.94921875,
|
92 |
+
"learning_rate": 4.7077603143418466e-05,
|
93 |
+
"loss": 0.0466,
|
94 |
+
"step": 120
|
95 |
+
},
|
96 |
+
{
|
97 |
+
"epoch": 0.06385068762278978,
|
98 |
+
"grad_norm": 0.90625,
|
99 |
+
"learning_rate": 4.683202357563851e-05,
|
100 |
+
"loss": 0.0472,
|
101 |
+
"step": 130
|
102 |
+
},
|
103 |
+
{
|
104 |
+
"epoch": 0.068762278978389,
|
105 |
+
"grad_norm": 1.1015625,
|
106 |
+
"learning_rate": 4.658644400785855e-05,
|
107 |
+
"loss": 0.0432,
|
108 |
+
"step": 140
|
109 |
+
},
|
110 |
+
{
|
111 |
+
"epoch": 0.07367387033398821,
|
112 |
+
"grad_norm": 1.1015625,
|
113 |
+
"learning_rate": 4.634086444007859e-05,
|
114 |
+
"loss": 0.0497,
|
115 |
+
"step": 150
|
116 |
+
},
|
117 |
+
{
|
118 |
+
"epoch": 0.07858546168958742,
|
119 |
+
"grad_norm": 1.3125,
|
120 |
+
"learning_rate": 4.609528487229863e-05,
|
121 |
+
"loss": 0.0406,
|
122 |
+
"step": 160
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 0.08349705304518663,
|
126 |
+
"grad_norm": 1.1015625,
|
127 |
+
"learning_rate": 4.584970530451866e-05,
|
128 |
+
"loss": 0.048,
|
129 |
+
"step": 170
|
130 |
+
},
|
131 |
+
{
|
132 |
+
"epoch": 0.08840864440078586,
|
133 |
+
"grad_norm": 0.875,
|
134 |
+
"learning_rate": 4.560412573673871e-05,
|
135 |
+
"loss": 0.0437,
|
136 |
+
"step": 180
|
137 |
+
},
|
138 |
+
{
|
139 |
+
"epoch": 0.09332023575638507,
|
140 |
+
"grad_norm": 0.8515625,
|
141 |
+
"learning_rate": 4.535854616895874e-05,
|
142 |
+
"loss": 0.0478,
|
143 |
+
"step": 190
|
144 |
+
},
|
145 |
+
{
|
146 |
+
"epoch": 0.09823182711198428,
|
147 |
+
"grad_norm": 0.8359375,
|
148 |
+
"learning_rate": 4.511296660117879e-05,
|
149 |
+
"loss": 0.0373,
|
150 |
+
"step": 200
|
151 |
+
},
|
152 |
+
{
|
153 |
+
"epoch": 0.1031434184675835,
|
154 |
+
"grad_norm": 0.6171875,
|
155 |
+
"learning_rate": 4.486738703339882e-05,
|
156 |
+
"loss": 0.0446,
|
157 |
+
"step": 210
|
158 |
+
},
|
159 |
+
{
|
160 |
+
"epoch": 0.10805500982318271,
|
161 |
+
"grad_norm": 0.81640625,
|
162 |
+
"learning_rate": 4.462180746561886e-05,
|
163 |
+
"loss": 0.0414,
|
164 |
+
"step": 220
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 0.11296660117878192,
|
168 |
+
"grad_norm": 0.70703125,
|
169 |
+
"learning_rate": 4.43762278978389e-05,
|
170 |
+
"loss": 0.0429,
|
171 |
+
"step": 230
|
172 |
+
},
|
173 |
+
{
|
174 |
+
"epoch": 0.11787819253438114,
|
175 |
+
"grad_norm": 1.046875,
|
176 |
+
"learning_rate": 4.413064833005894e-05,
|
177 |
+
"loss": 0.0386,
|
178 |
+
"step": 240
|
179 |
+
},
|
180 |
+
{
|
181 |
+
"epoch": 0.12278978388998035,
|
182 |
+
"grad_norm": 1.015625,
|
183 |
+
"learning_rate": 4.388506876227898e-05,
|
184 |
+
"loss": 0.0404,
|
185 |
+
"step": 250
|
186 |
+
},
|
187 |
+
{
|
188 |
+
"epoch": 0.12770137524557956,
|
189 |
+
"grad_norm": 1.0390625,
|
190 |
+
"learning_rate": 4.3639489194499023e-05,
|
191 |
+
"loss": 0.037,
|
192 |
+
"step": 260
|
193 |
+
},
|
194 |
+
{
|
195 |
+
"epoch": 0.13261296660117877,
|
196 |
+
"grad_norm": 0.9140625,
|
197 |
+
"learning_rate": 4.339390962671906e-05,
|
198 |
+
"loss": 0.0389,
|
199 |
+
"step": 270
|
200 |
+
},
|
201 |
+
{
|
202 |
+
"epoch": 0.137524557956778,
|
203 |
+
"grad_norm": 0.97265625,
|
204 |
+
"learning_rate": 4.31483300589391e-05,
|
205 |
+
"loss": 0.0416,
|
206 |
+
"step": 280
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 0.14243614931237722,
|
210 |
+
"grad_norm": 1.7421875,
|
211 |
+
"learning_rate": 4.290275049115914e-05,
|
212 |
+
"loss": 0.0377,
|
213 |
+
"step": 290
|
214 |
+
},
|
215 |
+
{
|
216 |
+
"epoch": 0.14734774066797643,
|
217 |
+
"grad_norm": 0.6015625,
|
218 |
+
"learning_rate": 4.265717092337918e-05,
|
219 |
+
"loss": 0.0327,
|
220 |
+
"step": 300
|
221 |
+
},
|
222 |
+
{
|
223 |
+
"epoch": 0.15225933202357564,
|
224 |
+
"grad_norm": 1.015625,
|
225 |
+
"learning_rate": 4.241159135559922e-05,
|
226 |
+
"loss": 0.0377,
|
227 |
+
"step": 310
|
228 |
+
},
|
229 |
+
{
|
230 |
+
"epoch": 0.15717092337917485,
|
231 |
+
"grad_norm": 0.87109375,
|
232 |
+
"learning_rate": 4.216601178781925e-05,
|
233 |
+
"loss": 0.038,
|
234 |
+
"step": 320
|
235 |
+
},
|
236 |
+
{
|
237 |
+
"epoch": 0.16208251473477406,
|
238 |
+
"grad_norm": 0.65625,
|
239 |
+
"learning_rate": 4.19204322200393e-05,
|
240 |
+
"loss": 0.0381,
|
241 |
+
"step": 330
|
242 |
+
},
|
243 |
+
{
|
244 |
+
"epoch": 0.16699410609037327,
|
245 |
+
"grad_norm": 0.6953125,
|
246 |
+
"learning_rate": 4.167485265225933e-05,
|
247 |
+
"loss": 0.0331,
|
248 |
+
"step": 340
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 0.1719056974459725,
|
252 |
+
"grad_norm": 0.76171875,
|
253 |
+
"learning_rate": 4.142927308447937e-05,
|
254 |
+
"loss": 0.0374,
|
255 |
+
"step": 350
|
256 |
+
},
|
257 |
+
{
|
258 |
+
"epoch": 0.17681728880157171,
|
259 |
+
"grad_norm": 0.90234375,
|
260 |
+
"learning_rate": 4.118369351669941e-05,
|
261 |
+
"loss": 0.0389,
|
262 |
+
"step": 360
|
263 |
+
},
|
264 |
+
{
|
265 |
+
"epoch": 0.18172888015717092,
|
266 |
+
"grad_norm": 1.1015625,
|
267 |
+
"learning_rate": 4.093811394891945e-05,
|
268 |
+
"loss": 0.0443,
|
269 |
+
"step": 370
|
270 |
+
},
|
271 |
+
{
|
272 |
+
"epoch": 0.18664047151277013,
|
273 |
+
"grad_norm": 0.6484375,
|
274 |
+
"learning_rate": 4.069253438113949e-05,
|
275 |
+
"loss": 0.0374,
|
276 |
+
"step": 380
|
277 |
+
},
|
278 |
+
{
|
279 |
+
"epoch": 0.19155206286836934,
|
280 |
+
"grad_norm": 0.47265625,
|
281 |
+
"learning_rate": 4.044695481335953e-05,
|
282 |
+
"loss": 0.0232,
|
283 |
+
"step": 390
|
284 |
+
},
|
285 |
+
{
|
286 |
+
"epoch": 0.19646365422396855,
|
287 |
+
"grad_norm": 0.92578125,
|
288 |
+
"learning_rate": 4.020137524557957e-05,
|
289 |
+
"loss": 0.043,
|
290 |
+
"step": 400
|
291 |
+
},
|
292 |
+
{
|
293 |
+
"epoch": 0.2013752455795678,
|
294 |
+
"grad_norm": 0.8125,
|
295 |
+
"learning_rate": 3.995579567779961e-05,
|
296 |
+
"loss": 0.0344,
|
297 |
+
"step": 410
|
298 |
+
},
|
299 |
+
{
|
300 |
+
"epoch": 0.206286836935167,
|
301 |
+
"grad_norm": 0.828125,
|
302 |
+
"learning_rate": 3.971021611001965e-05,
|
303 |
+
"loss": 0.0271,
|
304 |
+
"step": 420
|
305 |
+
},
|
306 |
+
{
|
307 |
+
"epoch": 0.2111984282907662,
|
308 |
+
"grad_norm": 0.953125,
|
309 |
+
"learning_rate": 3.946463654223969e-05,
|
310 |
+
"loss": 0.0378,
|
311 |
+
"step": 430
|
312 |
+
},
|
313 |
+
{
|
314 |
+
"epoch": 0.21611001964636542,
|
315 |
+
"grad_norm": 1.0,
|
316 |
+
"learning_rate": 3.921905697445973e-05,
|
317 |
+
"loss": 0.0356,
|
318 |
+
"step": 440
|
319 |
+
},
|
320 |
+
{
|
321 |
+
"epoch": 0.22102161100196463,
|
322 |
+
"grad_norm": 0.6953125,
|
323 |
+
"learning_rate": 3.897347740667976e-05,
|
324 |
+
"loss": 0.0336,
|
325 |
+
"step": 450
|
326 |
+
},
|
327 |
+
{
|
328 |
+
"epoch": 0.22593320235756384,
|
329 |
+
"grad_norm": 0.89453125,
|
330 |
+
"learning_rate": 3.872789783889981e-05,
|
331 |
+
"loss": 0.0334,
|
332 |
+
"step": 460
|
333 |
+
},
|
334 |
+
{
|
335 |
+
"epoch": 0.23084479371316308,
|
336 |
+
"grad_norm": 0.5390625,
|
337 |
+
"learning_rate": 3.848231827111984e-05,
|
338 |
+
"loss": 0.0316,
|
339 |
+
"step": 470
|
340 |
+
},
|
341 |
+
{
|
342 |
+
"epoch": 0.2357563850687623,
|
343 |
+
"grad_norm": 0.7421875,
|
344 |
+
"learning_rate": 3.823673870333989e-05,
|
345 |
+
"loss": 0.037,
|
346 |
+
"step": 480
|
347 |
+
},
|
348 |
+
{
|
349 |
+
"epoch": 0.2406679764243615,
|
350 |
+
"grad_norm": 0.65234375,
|
351 |
+
"learning_rate": 3.799115913555992e-05,
|
352 |
+
"loss": 0.0336,
|
353 |
+
"step": 490
|
354 |
+
},
|
355 |
+
{
|
356 |
+
"epoch": 0.2455795677799607,
|
357 |
+
"grad_norm": 0.67578125,
|
358 |
+
"learning_rate": 3.774557956777996e-05,
|
359 |
+
"loss": 0.0316,
|
360 |
+
"step": 500
|
361 |
+
},
|
362 |
+
{
|
363 |
+
"epoch": 0.2504911591355599,
|
364 |
+
"grad_norm": 0.89453125,
|
365 |
+
"learning_rate": 3.7500000000000003e-05,
|
366 |
+
"loss": 0.0287,
|
367 |
+
"step": 510
|
368 |
+
},
|
369 |
+
{
|
370 |
+
"epoch": 0.2554027504911591,
|
371 |
+
"grad_norm": 0.58984375,
|
372 |
+
"learning_rate": 3.725442043222004e-05,
|
373 |
+
"loss": 0.0353,
|
374 |
+
"step": 520
|
375 |
+
},
|
376 |
+
{
|
377 |
+
"epoch": 0.26031434184675833,
|
378 |
+
"grad_norm": 0.828125,
|
379 |
+
"learning_rate": 3.7008840864440084e-05,
|
380 |
+
"loss": 0.0323,
|
381 |
+
"step": 530
|
382 |
+
},
|
383 |
+
{
|
384 |
+
"epoch": 0.26522593320235754,
|
385 |
+
"grad_norm": 0.640625,
|
386 |
+
"learning_rate": 3.676326129666012e-05,
|
387 |
+
"loss": 0.0372,
|
388 |
+
"step": 540
|
389 |
+
},
|
390 |
+
{
|
391 |
+
"epoch": 0.27013752455795675,
|
392 |
+
"grad_norm": 0.8671875,
|
393 |
+
"learning_rate": 3.651768172888016e-05,
|
394 |
+
"loss": 0.0267,
|
395 |
+
"step": 550
|
396 |
+
},
|
397 |
+
{
|
398 |
+
"epoch": 0.275049115913556,
|
399 |
+
"grad_norm": 1.0859375,
|
400 |
+
"learning_rate": 3.62721021611002e-05,
|
401 |
+
"loss": 0.0304,
|
402 |
+
"step": 560
|
403 |
+
},
|
404 |
+
{
|
405 |
+
"epoch": 0.27996070726915523,
|
406 |
+
"grad_norm": 0.5859375,
|
407 |
+
"learning_rate": 3.602652259332024e-05,
|
408 |
+
"loss": 0.0326,
|
409 |
+
"step": 570
|
410 |
+
},
|
411 |
+
{
|
412 |
+
"epoch": 0.28487229862475444,
|
413 |
+
"grad_norm": 0.62890625,
|
414 |
+
"learning_rate": 3.578094302554028e-05,
|
415 |
+
"loss": 0.0403,
|
416 |
+
"step": 580
|
417 |
+
},
|
418 |
+
{
|
419 |
+
"epoch": 0.28978388998035365,
|
420 |
+
"grad_norm": 0.8515625,
|
421 |
+
"learning_rate": 3.553536345776032e-05,
|
422 |
+
"loss": 0.0306,
|
423 |
+
"step": 590
|
424 |
+
},
|
425 |
+
{
|
426 |
+
"epoch": 0.29469548133595286,
|
427 |
+
"grad_norm": 0.78515625,
|
428 |
+
"learning_rate": 3.528978388998035e-05,
|
429 |
+
"loss": 0.028,
|
430 |
+
"step": 600
|
431 |
+
},
|
432 |
+
{
|
433 |
+
"epoch": 0.29960707269155207,
|
434 |
+
"grad_norm": 0.578125,
|
435 |
+
"learning_rate": 3.50442043222004e-05,
|
436 |
+
"loss": 0.029,
|
437 |
+
"step": 610
|
438 |
+
},
|
439 |
+
{
|
440 |
+
"epoch": 0.3045186640471513,
|
441 |
+
"grad_norm": 0.6875,
|
442 |
+
"learning_rate": 3.479862475442043e-05,
|
443 |
+
"loss": 0.0299,
|
444 |
+
"step": 620
|
445 |
+
},
|
446 |
+
{
|
447 |
+
"epoch": 0.3094302554027505,
|
448 |
+
"grad_norm": 0.94140625,
|
449 |
+
"learning_rate": 3.455304518664047e-05,
|
450 |
+
"loss": 0.0293,
|
451 |
+
"step": 630
|
452 |
+
},
|
453 |
+
{
|
454 |
+
"epoch": 0.3143418467583497,
|
455 |
+
"grad_norm": 0.84375,
|
456 |
+
"learning_rate": 3.4307465618860514e-05,
|
457 |
+
"loss": 0.0288,
|
458 |
+
"step": 640
|
459 |
+
},
|
460 |
+
{
|
461 |
+
"epoch": 0.3192534381139489,
|
462 |
+
"grad_norm": 0.462890625,
|
463 |
+
"learning_rate": 3.406188605108055e-05,
|
464 |
+
"loss": 0.0272,
|
465 |
+
"step": 650
|
466 |
+
},
|
467 |
+
{
|
468 |
+
"epoch": 0.3241650294695481,
|
469 |
+
"grad_norm": 1.484375,
|
470 |
+
"learning_rate": 3.3816306483300594e-05,
|
471 |
+
"loss": 0.0392,
|
472 |
+
"step": 660
|
473 |
+
},
|
474 |
+
{
|
475 |
+
"epoch": 0.3290766208251473,
|
476 |
+
"grad_norm": 0.6640625,
|
477 |
+
"learning_rate": 3.357072691552063e-05,
|
478 |
+
"loss": 0.032,
|
479 |
+
"step": 670
|
480 |
+
},
|
481 |
+
{
|
482 |
+
"epoch": 0.33398821218074654,
|
483 |
+
"grad_norm": 1.109375,
|
484 |
+
"learning_rate": 3.332514734774067e-05,
|
485 |
+
"loss": 0.0388,
|
486 |
+
"step": 680
|
487 |
+
},
|
488 |
+
{
|
489 |
+
"epoch": 0.3388998035363458,
|
490 |
+
"grad_norm": 1.390625,
|
491 |
+
"learning_rate": 3.307956777996071e-05,
|
492 |
+
"loss": 0.03,
|
493 |
+
"step": 690
|
494 |
+
},
|
495 |
+
{
|
496 |
+
"epoch": 0.343811394891945,
|
497 |
+
"grad_norm": 0.80859375,
|
498 |
+
"learning_rate": 3.283398821218075e-05,
|
499 |
+
"loss": 0.0316,
|
500 |
+
"step": 700
|
501 |
+
},
|
502 |
+
{
|
503 |
+
"epoch": 0.3487229862475442,
|
504 |
+
"grad_norm": 0.6796875,
|
505 |
+
"learning_rate": 3.258840864440079e-05,
|
506 |
+
"loss": 0.037,
|
507 |
+
"step": 710
|
508 |
+
},
|
509 |
+
{
|
510 |
+
"epoch": 0.35363457760314343,
|
511 |
+
"grad_norm": 0.78125,
|
512 |
+
"learning_rate": 3.234282907662083e-05,
|
513 |
+
"loss": 0.0302,
|
514 |
+
"step": 720
|
515 |
+
},
|
516 |
+
{
|
517 |
+
"epoch": 0.35854616895874264,
|
518 |
+
"grad_norm": 0.5546875,
|
519 |
+
"learning_rate": 3.209724950884086e-05,
|
520 |
+
"loss": 0.0294,
|
521 |
+
"step": 730
|
522 |
+
},
|
523 |
+
{
|
524 |
+
"epoch": 0.36345776031434185,
|
525 |
+
"grad_norm": 0.60546875,
|
526 |
+
"learning_rate": 3.185166994106091e-05,
|
527 |
+
"loss": 0.0303,
|
528 |
+
"step": 740
|
529 |
+
},
|
530 |
+
{
|
531 |
+
"epoch": 0.36836935166994106,
|
532 |
+
"grad_norm": 1.125,
|
533 |
+
"learning_rate": 3.160609037328094e-05,
|
534 |
+
"loss": 0.0304,
|
535 |
+
"step": 750
|
536 |
+
},
|
537 |
+
{
|
538 |
+
"epoch": 0.37328094302554027,
|
539 |
+
"grad_norm": 0.609375,
|
540 |
+
"learning_rate": 3.1360510805500984e-05,
|
541 |
+
"loss": 0.0303,
|
542 |
+
"step": 760
|
543 |
+
},
|
544 |
+
{
|
545 |
+
"epoch": 0.3781925343811395,
|
546 |
+
"grad_norm": 0.9609375,
|
547 |
+
"learning_rate": 3.1114931237721024e-05,
|
548 |
+
"loss": 0.0323,
|
549 |
+
"step": 770
|
550 |
+
},
|
551 |
+
{
|
552 |
+
"epoch": 0.3831041257367387,
|
553 |
+
"grad_norm": 0.796875,
|
554 |
+
"learning_rate": 3.086935166994106e-05,
|
555 |
+
"loss": 0.0324,
|
556 |
+
"step": 780
|
557 |
+
},
|
558 |
+
{
|
559 |
+
"epoch": 0.3880157170923379,
|
560 |
+
"grad_norm": 0.7421875,
|
561 |
+
"learning_rate": 3.0623772102161104e-05,
|
562 |
+
"loss": 0.0318,
|
563 |
+
"step": 790
|
564 |
+
},
|
565 |
+
{
|
566 |
+
"epoch": 0.3929273084479371,
|
567 |
+
"grad_norm": 0.6484375,
|
568 |
+
"learning_rate": 3.0378192534381138e-05,
|
569 |
+
"loss": 0.0295,
|
570 |
+
"step": 800
|
571 |
+
},
|
572 |
+
{
|
573 |
+
"epoch": 0.39783889980353637,
|
574 |
+
"grad_norm": 0.6484375,
|
575 |
+
"learning_rate": 3.013261296660118e-05,
|
576 |
+
"loss": 0.0267,
|
577 |
+
"step": 810
|
578 |
+
},
|
579 |
+
{
|
580 |
+
"epoch": 0.4027504911591356,
|
581 |
+
"grad_norm": 1.0234375,
|
582 |
+
"learning_rate": 2.988703339882122e-05,
|
583 |
+
"loss": 0.0324,
|
584 |
+
"step": 820
|
585 |
+
},
|
586 |
+
{
|
587 |
+
"epoch": 0.4076620825147348,
|
588 |
+
"grad_norm": 0.81640625,
|
589 |
+
"learning_rate": 2.964145383104126e-05,
|
590 |
+
"loss": 0.027,
|
591 |
+
"step": 830
|
592 |
+
},
|
593 |
+
{
|
594 |
+
"epoch": 0.412573673870334,
|
595 |
+
"grad_norm": 0.609375,
|
596 |
+
"learning_rate": 2.9395874263261296e-05,
|
597 |
+
"loss": 0.0294,
|
598 |
+
"step": 840
|
599 |
+
},
|
600 |
+
{
|
601 |
+
"epoch": 0.4174852652259332,
|
602 |
+
"grad_norm": 0.625,
|
603 |
+
"learning_rate": 2.915029469548134e-05,
|
604 |
+
"loss": 0.0265,
|
605 |
+
"step": 850
|
606 |
+
},
|
607 |
+
{
|
608 |
+
"epoch": 0.4223968565815324,
|
609 |
+
"grad_norm": 0.6640625,
|
610 |
+
"learning_rate": 2.8904715127701376e-05,
|
611 |
+
"loss": 0.027,
|
612 |
+
"step": 860
|
613 |
+
},
|
614 |
+
{
|
615 |
+
"epoch": 0.42730844793713163,
|
616 |
+
"grad_norm": 0.4921875,
|
617 |
+
"learning_rate": 2.865913555992142e-05,
|
618 |
+
"loss": 0.0308,
|
619 |
+
"step": 870
|
620 |
+
},
|
621 |
+
{
|
622 |
+
"epoch": 0.43222003929273084,
|
623 |
+
"grad_norm": 0.57421875,
|
624 |
+
"learning_rate": 2.8413555992141457e-05,
|
625 |
+
"loss": 0.0312,
|
626 |
+
"step": 880
|
627 |
+
},
|
628 |
+
{
|
629 |
+
"epoch": 0.43713163064833005,
|
630 |
+
"grad_norm": 0.8984375,
|
631 |
+
"learning_rate": 2.816797642436149e-05,
|
632 |
+
"loss": 0.0297,
|
633 |
+
"step": 890
|
634 |
+
},
|
635 |
+
{
|
636 |
+
"epoch": 0.44204322200392926,
|
637 |
+
"grad_norm": 0.5703125,
|
638 |
+
"learning_rate": 2.7922396856581534e-05,
|
639 |
+
"loss": 0.0308,
|
640 |
+
"step": 900
|
641 |
+
},
|
642 |
+
{
|
643 |
+
"epoch": 0.44695481335952847,
|
644 |
+
"grad_norm": 0.8203125,
|
645 |
+
"learning_rate": 2.767681728880157e-05,
|
646 |
+
"loss": 0.0307,
|
647 |
+
"step": 910
|
648 |
+
},
|
649 |
+
{
|
650 |
+
"epoch": 0.4518664047151277,
|
651 |
+
"grad_norm": 0.6015625,
|
652 |
+
"learning_rate": 2.7431237721021615e-05,
|
653 |
+
"loss": 0.031,
|
654 |
+
"step": 920
|
655 |
+
},
|
656 |
+
{
|
657 |
+
"epoch": 0.4567779960707269,
|
658 |
+
"grad_norm": 0.609375,
|
659 |
+
"learning_rate": 2.718565815324165e-05,
|
660 |
+
"loss": 0.0388,
|
661 |
+
"step": 930
|
662 |
+
},
|
663 |
+
{
|
664 |
+
"epoch": 0.46168958742632615,
|
665 |
+
"grad_norm": 0.578125,
|
666 |
+
"learning_rate": 2.6940078585461692e-05,
|
667 |
+
"loss": 0.0293,
|
668 |
+
"step": 940
|
669 |
+
},
|
670 |
+
{
|
671 |
+
"epoch": 0.46660117878192536,
|
672 |
+
"grad_norm": 0.90234375,
|
673 |
+
"learning_rate": 2.669449901768173e-05,
|
674 |
+
"loss": 0.0339,
|
675 |
+
"step": 950
|
676 |
+
},
|
677 |
+
{
|
678 |
+
"epoch": 0.4715127701375246,
|
679 |
+
"grad_norm": 0.5625,
|
680 |
+
"learning_rate": 2.6448919449901772e-05,
|
681 |
+
"loss": 0.0322,
|
682 |
+
"step": 960
|
683 |
+
},
|
684 |
+
{
|
685 |
+
"epoch": 0.4764243614931238,
|
686 |
+
"grad_norm": 0.5234375,
|
687 |
+
"learning_rate": 2.620333988212181e-05,
|
688 |
+
"loss": 0.0316,
|
689 |
+
"step": 970
|
690 |
+
},
|
691 |
+
{
|
692 |
+
"epoch": 0.481335952848723,
|
693 |
+
"grad_norm": 0.7734375,
|
694 |
+
"learning_rate": 2.595776031434185e-05,
|
695 |
+
"loss": 0.0329,
|
696 |
+
"step": 980
|
697 |
+
},
|
698 |
+
{
|
699 |
+
"epoch": 0.4862475442043222,
|
700 |
+
"grad_norm": 0.58984375,
|
701 |
+
"learning_rate": 2.5712180746561886e-05,
|
702 |
+
"loss": 0.0251,
|
703 |
+
"step": 990
|
704 |
+
},
|
705 |
+
{
|
706 |
+
"epoch": 0.4911591355599214,
|
707 |
+
"grad_norm": 0.9921875,
|
708 |
+
"learning_rate": 2.546660117878193e-05,
|
709 |
+
"loss": 0.0256,
|
710 |
+
"step": 1000
|
711 |
+
},
|
712 |
+
{
|
713 |
+
"epoch": 0.4960707269155206,
|
714 |
+
"grad_norm": 1.1015625,
|
715 |
+
"learning_rate": 2.5221021611001967e-05,
|
716 |
+
"loss": 0.0294,
|
717 |
+
"step": 1010
|
718 |
+
},
|
719 |
+
{
|
720 |
+
"epoch": 0.5009823182711198,
|
721 |
+
"grad_norm": 0.6484375,
|
722 |
+
"learning_rate": 2.4975442043222004e-05,
|
723 |
+
"loss": 0.0251,
|
724 |
+
"step": 1020
|
725 |
+
},
|
726 |
+
{
|
727 |
+
"epoch": 0.5058939096267191,
|
728 |
+
"grad_norm": 0.55859375,
|
729 |
+
"learning_rate": 2.4729862475442044e-05,
|
730 |
+
"loss": 0.0275,
|
731 |
+
"step": 1030
|
732 |
+
},
|
733 |
+
{
|
734 |
+
"epoch": 0.5108055009823183,
|
735 |
+
"grad_norm": 0.5546875,
|
736 |
+
"learning_rate": 2.4484282907662084e-05,
|
737 |
+
"loss": 0.0261,
|
738 |
+
"step": 1040
|
739 |
+
},
|
740 |
+
{
|
741 |
+
"epoch": 0.5157170923379175,
|
742 |
+
"grad_norm": 0.72265625,
|
743 |
+
"learning_rate": 2.4238703339882125e-05,
|
744 |
+
"loss": 0.0292,
|
745 |
+
"step": 1050
|
746 |
+
},
|
747 |
+
{
|
748 |
+
"epoch": 0.5206286836935167,
|
749 |
+
"grad_norm": 0.640625,
|
750 |
+
"learning_rate": 2.3993123772102165e-05,
|
751 |
+
"loss": 0.0252,
|
752 |
+
"step": 1060
|
753 |
+
},
|
754 |
+
{
|
755 |
+
"epoch": 0.5255402750491159,
|
756 |
+
"grad_norm": 0.671875,
|
757 |
+
"learning_rate": 2.37475442043222e-05,
|
758 |
+
"loss": 0.0305,
|
759 |
+
"step": 1070
|
760 |
+
},
|
761 |
+
{
|
762 |
+
"epoch": 0.5304518664047151,
|
763 |
+
"grad_norm": 0.74609375,
|
764 |
+
"learning_rate": 2.350196463654224e-05,
|
765 |
+
"loss": 0.029,
|
766 |
+
"step": 1080
|
767 |
+
},
|
768 |
+
{
|
769 |
+
"epoch": 0.5353634577603144,
|
770 |
+
"grad_norm": 0.62109375,
|
771 |
+
"learning_rate": 2.325638506876228e-05,
|
772 |
+
"loss": 0.0306,
|
773 |
+
"step": 1090
|
774 |
+
},
|
775 |
+
{
|
776 |
+
"epoch": 0.5402750491159135,
|
777 |
+
"grad_norm": 0.5078125,
|
778 |
+
"learning_rate": 2.301080550098232e-05,
|
779 |
+
"loss": 0.0295,
|
780 |
+
"step": 1100
|
781 |
+
},
|
782 |
+
{
|
783 |
+
"epoch": 0.5451866404715128,
|
784 |
+
"grad_norm": 0.55859375,
|
785 |
+
"learning_rate": 2.276522593320236e-05,
|
786 |
+
"loss": 0.0244,
|
787 |
+
"step": 1110
|
788 |
+
},
|
789 |
+
{
|
790 |
+
"epoch": 0.550098231827112,
|
791 |
+
"grad_norm": 0.7890625,
|
792 |
+
"learning_rate": 2.2519646365422397e-05,
|
793 |
+
"loss": 0.0261,
|
794 |
+
"step": 1120
|
795 |
+
},
|
796 |
+
{
|
797 |
+
"epoch": 0.5550098231827112,
|
798 |
+
"grad_norm": 0.59375,
|
799 |
+
"learning_rate": 2.2274066797642437e-05,
|
800 |
+
"loss": 0.0227,
|
801 |
+
"step": 1130
|
802 |
+
},
|
803 |
+
{
|
804 |
+
"epoch": 0.5599214145383105,
|
805 |
+
"grad_norm": 0.44140625,
|
806 |
+
"learning_rate": 2.2028487229862477e-05,
|
807 |
+
"loss": 0.0244,
|
808 |
+
"step": 1140
|
809 |
+
},
|
810 |
+
{
|
811 |
+
"epoch": 0.5648330058939096,
|
812 |
+
"grad_norm": 0.63671875,
|
813 |
+
"learning_rate": 2.1782907662082517e-05,
|
814 |
+
"loss": 0.0235,
|
815 |
+
"step": 1150
|
816 |
+
},
|
817 |
+
{
|
818 |
+
"epoch": 0.5697445972495089,
|
819 |
+
"grad_norm": 0.60546875,
|
820 |
+
"learning_rate": 2.1537328094302554e-05,
|
821 |
+
"loss": 0.0297,
|
822 |
+
"step": 1160
|
823 |
+
},
|
824 |
+
{
|
825 |
+
"epoch": 0.574656188605108,
|
826 |
+
"grad_norm": 0.70703125,
|
827 |
+
"learning_rate": 2.1291748526522595e-05,
|
828 |
+
"loss": 0.0267,
|
829 |
+
"step": 1170
|
830 |
+
},
|
831 |
+
{
|
832 |
+
"epoch": 0.5795677799607073,
|
833 |
+
"grad_norm": 1.03125,
|
834 |
+
"learning_rate": 2.1046168958742635e-05,
|
835 |
+
"loss": 0.0276,
|
836 |
+
"step": 1180
|
837 |
+
},
|
838 |
+
{
|
839 |
+
"epoch": 0.5844793713163065,
|
840 |
+
"grad_norm": 0.5390625,
|
841 |
+
"learning_rate": 2.0800589390962675e-05,
|
842 |
+
"loss": 0.0281,
|
843 |
+
"step": 1190
|
844 |
+
},
|
845 |
+
{
|
846 |
+
"epoch": 0.5893909626719057,
|
847 |
+
"grad_norm": 0.6796875,
|
848 |
+
"learning_rate": 2.0555009823182712e-05,
|
849 |
+
"loss": 0.0306,
|
850 |
+
"step": 1200
|
851 |
+
},
|
852 |
+
{
|
853 |
+
"epoch": 0.5943025540275049,
|
854 |
+
"grad_norm": 0.46875,
|
855 |
+
"learning_rate": 2.030943025540275e-05,
|
856 |
+
"loss": 0.0253,
|
857 |
+
"step": 1210
|
858 |
+
},
|
859 |
+
{
|
860 |
+
"epoch": 0.5992141453831041,
|
861 |
+
"grad_norm": 0.625,
|
862 |
+
"learning_rate": 2.006385068762279e-05,
|
863 |
+
"loss": 0.0258,
|
864 |
+
"step": 1220
|
865 |
+
},
|
866 |
+
{
|
867 |
+
"epoch": 0.6041257367387033,
|
868 |
+
"grad_norm": 0.67578125,
|
869 |
+
"learning_rate": 1.981827111984283e-05,
|
870 |
+
"loss": 0.0275,
|
871 |
+
"step": 1230
|
872 |
+
},
|
873 |
+
{
|
874 |
+
"epoch": 0.6090373280943026,
|
875 |
+
"grad_norm": 0.78515625,
|
876 |
+
"learning_rate": 1.957269155206287e-05,
|
877 |
+
"loss": 0.0275,
|
878 |
+
"step": 1240
|
879 |
+
},
|
880 |
+
{
|
881 |
+
"epoch": 0.6139489194499018,
|
882 |
+
"grad_norm": 0.8125,
|
883 |
+
"learning_rate": 1.932711198428291e-05,
|
884 |
+
"loss": 0.0304,
|
885 |
+
"step": 1250
|
886 |
+
},
|
887 |
+
{
|
888 |
+
"epoch": 0.618860510805501,
|
889 |
+
"grad_norm": 0.703125,
|
890 |
+
"learning_rate": 1.9081532416502947e-05,
|
891 |
+
"loss": 0.0234,
|
892 |
+
"step": 1260
|
893 |
+
},
|
894 |
+
{
|
895 |
+
"epoch": 0.6237721021611002,
|
896 |
+
"grad_norm": 0.77734375,
|
897 |
+
"learning_rate": 1.8835952848722987e-05,
|
898 |
+
"loss": 0.0315,
|
899 |
+
"step": 1270
|
900 |
+
},
|
901 |
+
{
|
902 |
+
"epoch": 0.6286836935166994,
|
903 |
+
"grad_norm": 0.82421875,
|
904 |
+
"learning_rate": 1.8590373280943028e-05,
|
905 |
+
"loss": 0.0293,
|
906 |
+
"step": 1280
|
907 |
+
},
|
908 |
+
{
|
909 |
+
"epoch": 0.6335952848722987,
|
910 |
+
"grad_norm": 0.4609375,
|
911 |
+
"learning_rate": 1.8344793713163068e-05,
|
912 |
+
"loss": 0.0261,
|
913 |
+
"step": 1290
|
914 |
+
},
|
915 |
+
{
|
916 |
+
"epoch": 0.6385068762278978,
|
917 |
+
"grad_norm": 0.85546875,
|
918 |
+
"learning_rate": 1.8099214145383105e-05,
|
919 |
+
"loss": 0.0264,
|
920 |
+
"step": 1300
|
921 |
+
},
|
922 |
+
{
|
923 |
+
"epoch": 0.6434184675834971,
|
924 |
+
"grad_norm": 0.73828125,
|
925 |
+
"learning_rate": 1.7853634577603145e-05,
|
926 |
+
"loss": 0.0283,
|
927 |
+
"step": 1310
|
928 |
+
},
|
929 |
+
{
|
930 |
+
"epoch": 0.6483300589390962,
|
931 |
+
"grad_norm": 1.0625,
|
932 |
+
"learning_rate": 1.7608055009823182e-05,
|
933 |
+
"loss": 0.0244,
|
934 |
+
"step": 1320
|
935 |
+
},
|
936 |
+
{
|
937 |
+
"epoch": 0.6532416502946955,
|
938 |
+
"grad_norm": 0.8203125,
|
939 |
+
"learning_rate": 1.7362475442043222e-05,
|
940 |
+
"loss": 0.0249,
|
941 |
+
"step": 1330
|
942 |
+
},
|
943 |
+
{
|
944 |
+
"epoch": 0.6581532416502947,
|
945 |
+
"grad_norm": 0.765625,
|
946 |
+
"learning_rate": 1.7116895874263263e-05,
|
947 |
+
"loss": 0.0246,
|
948 |
+
"step": 1340
|
949 |
+
},
|
950 |
+
{
|
951 |
+
"epoch": 0.6630648330058939,
|
952 |
+
"grad_norm": 0.77734375,
|
953 |
+
"learning_rate": 1.68713163064833e-05,
|
954 |
+
"loss": 0.0268,
|
955 |
+
"step": 1350
|
956 |
+
},
|
957 |
+
{
|
958 |
+
"epoch": 0.6679764243614931,
|
959 |
+
"grad_norm": 0.5390625,
|
960 |
+
"learning_rate": 1.662573673870334e-05,
|
961 |
+
"loss": 0.025,
|
962 |
+
"step": 1360
|
963 |
+
},
|
964 |
+
{
|
965 |
+
"epoch": 0.6728880157170923,
|
966 |
+
"grad_norm": 0.82421875,
|
967 |
+
"learning_rate": 1.638015717092338e-05,
|
968 |
+
"loss": 0.0209,
|
969 |
+
"step": 1370
|
970 |
+
},
|
971 |
+
{
|
972 |
+
"epoch": 0.6777996070726916,
|
973 |
+
"grad_norm": 0.5546875,
|
974 |
+
"learning_rate": 1.613457760314342e-05,
|
975 |
+
"loss": 0.0264,
|
976 |
+
"step": 1380
|
977 |
+
},
|
978 |
+
{
|
979 |
+
"epoch": 0.6827111984282908,
|
980 |
+
"grad_norm": 1.0703125,
|
981 |
+
"learning_rate": 1.5888998035363457e-05,
|
982 |
+
"loss": 0.0227,
|
983 |
+
"step": 1390
|
984 |
+
},
|
985 |
+
{
|
986 |
+
"epoch": 0.68762278978389,
|
987 |
+
"grad_norm": 0.66015625,
|
988 |
+
"learning_rate": 1.5643418467583497e-05,
|
989 |
+
"loss": 0.0313,
|
990 |
+
"step": 1400
|
991 |
+
},
|
992 |
+
{
|
993 |
+
"epoch": 0.6925343811394892,
|
994 |
+
"grad_norm": 0.7578125,
|
995 |
+
"learning_rate": 1.5397838899803538e-05,
|
996 |
+
"loss": 0.0242,
|
997 |
+
"step": 1410
|
998 |
+
},
|
999 |
+
{
|
1000 |
+
"epoch": 0.6974459724950884,
|
1001 |
+
"grad_norm": 0.6796875,
|
1002 |
+
"learning_rate": 1.5152259332023578e-05,
|
1003 |
+
"loss": 0.0274,
|
1004 |
+
"step": 1420
|
1005 |
+
},
|
1006 |
+
{
|
1007 |
+
"epoch": 0.7023575638506876,
|
1008 |
+
"grad_norm": 0.80078125,
|
1009 |
+
"learning_rate": 1.4906679764243617e-05,
|
1010 |
+
"loss": 0.0235,
|
1011 |
+
"step": 1430
|
1012 |
+
},
|
1013 |
+
{
|
1014 |
+
"epoch": 0.7072691552062869,
|
1015 |
+
"grad_norm": 0.66015625,
|
1016 |
+
"learning_rate": 1.4661100196463657e-05,
|
1017 |
+
"loss": 0.0259,
|
1018 |
+
"step": 1440
|
1019 |
+
},
|
1020 |
+
{
|
1021 |
+
"epoch": 0.712180746561886,
|
1022 |
+
"grad_norm": 0.453125,
|
1023 |
+
"learning_rate": 1.4415520628683694e-05,
|
1024 |
+
"loss": 0.0246,
|
1025 |
+
"step": 1450
|
1026 |
+
},
|
1027 |
+
{
|
1028 |
+
"epoch": 0.7170923379174853,
|
1029 |
+
"grad_norm": 0.71484375,
|
1030 |
+
"learning_rate": 1.4169941060903732e-05,
|
1031 |
+
"loss": 0.0275,
|
1032 |
+
"step": 1460
|
1033 |
+
},
|
1034 |
+
{
|
1035 |
+
"epoch": 0.7220039292730844,
|
1036 |
+
"grad_norm": 1.1015625,
|
1037 |
+
"learning_rate": 1.3924361493123773e-05,
|
1038 |
+
"loss": 0.0308,
|
1039 |
+
"step": 1470
|
1040 |
+
},
|
1041 |
+
{
|
1042 |
+
"epoch": 0.7269155206286837,
|
1043 |
+
"grad_norm": 0.9765625,
|
1044 |
+
"learning_rate": 1.3678781925343811e-05,
|
1045 |
+
"loss": 0.029,
|
1046 |
+
"step": 1480
|
1047 |
+
},
|
1048 |
+
{
|
1049 |
+
"epoch": 0.731827111984283,
|
1050 |
+
"grad_norm": 1.7578125,
|
1051 |
+
"learning_rate": 1.3433202357563852e-05,
|
1052 |
+
"loss": 0.0284,
|
1053 |
+
"step": 1490
|
1054 |
+
},
|
1055 |
+
{
|
1056 |
+
"epoch": 0.7367387033398821,
|
1057 |
+
"grad_norm": 0.7265625,
|
1058 |
+
"learning_rate": 1.318762278978389e-05,
|
1059 |
+
"loss": 0.0253,
|
1060 |
+
"step": 1500
|
1061 |
+
},
|
1062 |
+
{
|
1063 |
+
"epoch": 0.7416502946954814,
|
1064 |
+
"grad_norm": 0.8515625,
|
1065 |
+
"learning_rate": 1.294204322200393e-05,
|
1066 |
+
"loss": 0.028,
|
1067 |
+
"step": 1510
|
1068 |
+
},
|
1069 |
+
{
|
1070 |
+
"epoch": 0.7465618860510805,
|
1071 |
+
"grad_norm": 0.53515625,
|
1072 |
+
"learning_rate": 1.2696463654223969e-05,
|
1073 |
+
"loss": 0.0292,
|
1074 |
+
"step": 1520
|
1075 |
+
},
|
1076 |
+
{
|
1077 |
+
"epoch": 0.7514734774066798,
|
1078 |
+
"grad_norm": 0.62890625,
|
1079 |
+
"learning_rate": 1.245088408644401e-05,
|
1080 |
+
"loss": 0.0259,
|
1081 |
+
"step": 1530
|
1082 |
+
},
|
1083 |
+
{
|
1084 |
+
"epoch": 0.756385068762279,
|
1085 |
+
"grad_norm": 0.6171875,
|
1086 |
+
"learning_rate": 1.2205304518664048e-05,
|
1087 |
+
"loss": 0.0263,
|
1088 |
+
"step": 1540
|
1089 |
+
},
|
1090 |
+
{
|
1091 |
+
"epoch": 0.7612966601178782,
|
1092 |
+
"grad_norm": 0.72265625,
|
1093 |
+
"learning_rate": 1.1959724950884087e-05,
|
1094 |
+
"loss": 0.0279,
|
1095 |
+
"step": 1550
|
1096 |
+
},
|
1097 |
+
{
|
1098 |
+
"epoch": 0.7662082514734774,
|
1099 |
+
"grad_norm": 0.70703125,
|
1100 |
+
"learning_rate": 1.1714145383104127e-05,
|
1101 |
+
"loss": 0.0276,
|
1102 |
+
"step": 1560
|
1103 |
+
},
|
1104 |
+
{
|
1105 |
+
"epoch": 0.7711198428290766,
|
1106 |
+
"grad_norm": 0.98828125,
|
1107 |
+
"learning_rate": 1.1468565815324165e-05,
|
1108 |
+
"loss": 0.0314,
|
1109 |
+
"step": 1570
|
1110 |
+
},
|
1111 |
+
{
|
1112 |
+
"epoch": 0.7760314341846758,
|
1113 |
+
"grad_norm": 0.6640625,
|
1114 |
+
"learning_rate": 1.1222986247544206e-05,
|
1115 |
+
"loss": 0.0283,
|
1116 |
+
"step": 1580
|
1117 |
+
},
|
1118 |
+
{
|
1119 |
+
"epoch": 0.7809430255402751,
|
1120 |
+
"grad_norm": 0.703125,
|
1121 |
+
"learning_rate": 1.0977406679764244e-05,
|
1122 |
+
"loss": 0.049,
|
1123 |
+
"step": 1590
|
1124 |
+
},
|
1125 |
+
{
|
1126 |
+
"epoch": 0.7858546168958742,
|
1127 |
+
"grad_norm": 0.73828125,
|
1128 |
+
"learning_rate": 1.0731827111984283e-05,
|
1129 |
+
"loss": 0.0294,
|
1130 |
+
"step": 1600
|
1131 |
+
},
|
1132 |
+
{
|
1133 |
+
"epoch": 0.7907662082514735,
|
1134 |
+
"grad_norm": 0.78125,
|
1135 |
+
"learning_rate": 1.0486247544204323e-05,
|
1136 |
+
"loss": 0.0257,
|
1137 |
+
"step": 1610
|
1138 |
+
},
|
1139 |
+
{
|
1140 |
+
"epoch": 0.7956777996070727,
|
1141 |
+
"grad_norm": 0.69140625,
|
1142 |
+
"learning_rate": 1.0240667976424362e-05,
|
1143 |
+
"loss": 0.0316,
|
1144 |
+
"step": 1620
|
1145 |
+
},
|
1146 |
+
{
|
1147 |
+
"epoch": 0.8005893909626719,
|
1148 |
+
"grad_norm": 1.0234375,
|
1149 |
+
"learning_rate": 9.995088408644402e-06,
|
1150 |
+
"loss": 0.033,
|
1151 |
+
"step": 1630
|
1152 |
+
},
|
1153 |
+
{
|
1154 |
+
"epoch": 0.8055009823182712,
|
1155 |
+
"grad_norm": 0.77734375,
|
1156 |
+
"learning_rate": 9.74950884086444e-06,
|
1157 |
+
"loss": 0.0263,
|
1158 |
+
"step": 1640
|
1159 |
+
},
|
1160 |
+
{
|
1161 |
+
"epoch": 0.8104125736738703,
|
1162 |
+
"grad_norm": 0.69140625,
|
1163 |
+
"learning_rate": 9.503929273084481e-06,
|
1164 |
+
"loss": 0.0248,
|
1165 |
+
"step": 1650
|
1166 |
+
},
|
1167 |
+
{
|
1168 |
+
"epoch": 0.8153241650294696,
|
1169 |
+
"grad_norm": 0.80859375,
|
1170 |
+
"learning_rate": 9.258349705304518e-06,
|
1171 |
+
"loss": 0.0256,
|
1172 |
+
"step": 1660
|
1173 |
+
},
|
1174 |
+
{
|
1175 |
+
"epoch": 0.8202357563850687,
|
1176 |
+
"grad_norm": 0.498046875,
|
1177 |
+
"learning_rate": 9.012770137524558e-06,
|
1178 |
+
"loss": 0.0258,
|
1179 |
+
"step": 1670
|
1180 |
+
},
|
1181 |
+
{
|
1182 |
+
"epoch": 0.825147347740668,
|
1183 |
+
"grad_norm": 0.64453125,
|
1184 |
+
"learning_rate": 8.767190569744597e-06,
|
1185 |
+
"loss": 0.0271,
|
1186 |
+
"step": 1680
|
1187 |
+
},
|
1188 |
+
{
|
1189 |
+
"epoch": 0.8300589390962672,
|
1190 |
+
"grad_norm": 0.7890625,
|
1191 |
+
"learning_rate": 8.521611001964637e-06,
|
1192 |
+
"loss": 0.0334,
|
1193 |
+
"step": 1690
|
1194 |
+
},
|
1195 |
+
{
|
1196 |
+
"epoch": 0.8349705304518664,
|
1197 |
+
"grad_norm": 1.0078125,
|
1198 |
+
"learning_rate": 8.276031434184677e-06,
|
1199 |
+
"loss": 0.0245,
|
1200 |
+
"step": 1700
|
1201 |
+
},
|
1202 |
+
{
|
1203 |
+
"epoch": 0.8398821218074656,
|
1204 |
+
"grad_norm": 0.63671875,
|
1205 |
+
"learning_rate": 8.030451866404716e-06,
|
1206 |
+
"loss": 0.0262,
|
1207 |
+
"step": 1710
|
1208 |
+
},
|
1209 |
+
{
|
1210 |
+
"epoch": 0.8447937131630648,
|
1211 |
+
"grad_norm": 0.5390625,
|
1212 |
+
"learning_rate": 7.784872298624756e-06,
|
1213 |
+
"loss": 0.027,
|
1214 |
+
"step": 1720
|
1215 |
+
},
|
1216 |
+
{
|
1217 |
+
"epoch": 0.849705304518664,
|
1218 |
+
"grad_norm": 0.6953125,
|
1219 |
+
"learning_rate": 7.539292730844794e-06,
|
1220 |
+
"loss": 0.0253,
|
1221 |
+
"step": 1730
|
1222 |
+
},
|
1223 |
+
{
|
1224 |
+
"epoch": 0.8546168958742633,
|
1225 |
+
"grad_norm": 0.62109375,
|
1226 |
+
"learning_rate": 7.293713163064833e-06,
|
1227 |
+
"loss": 0.0247,
|
1228 |
+
"step": 1740
|
1229 |
+
},
|
1230 |
+
{
|
1231 |
+
"epoch": 0.8595284872298625,
|
1232 |
+
"grad_norm": 1.140625,
|
1233 |
+
"learning_rate": 7.048133595284873e-06,
|
1234 |
+
"loss": 0.0281,
|
1235 |
+
"step": 1750
|
1236 |
+
},
|
1237 |
+
{
|
1238 |
+
"epoch": 0.8644400785854617,
|
1239 |
+
"grad_norm": 0.59375,
|
1240 |
+
"learning_rate": 6.802554027504912e-06,
|
1241 |
+
"loss": 0.0269,
|
1242 |
+
"step": 1760
|
1243 |
+
},
|
1244 |
+
{
|
1245 |
+
"epoch": 0.869351669941061,
|
1246 |
+
"grad_norm": 1.2421875,
|
1247 |
+
"learning_rate": 6.556974459724952e-06,
|
1248 |
+
"loss": 0.0255,
|
1249 |
+
"step": 1770
|
1250 |
+
},
|
1251 |
+
{
|
1252 |
+
"epoch": 0.8742632612966601,
|
1253 |
+
"grad_norm": 0.7578125,
|
1254 |
+
"learning_rate": 6.311394891944991e-06,
|
1255 |
+
"loss": 0.0253,
|
1256 |
+
"step": 1780
|
1257 |
+
},
|
1258 |
+
{
|
1259 |
+
"epoch": 0.8791748526522594,
|
1260 |
+
"grad_norm": 0.73828125,
|
1261 |
+
"learning_rate": 6.06581532416503e-06,
|
1262 |
+
"loss": 0.024,
|
1263 |
+
"step": 1790
|
1264 |
+
},
|
1265 |
+
{
|
1266 |
+
"epoch": 0.8840864440078585,
|
1267 |
+
"grad_norm": 0.60546875,
|
1268 |
+
"learning_rate": 5.820235756385069e-06,
|
1269 |
+
"loss": 0.03,
|
1270 |
+
"step": 1800
|
1271 |
+
},
|
1272 |
+
{
|
1273 |
+
"epoch": 0.8889980353634578,
|
1274 |
+
"grad_norm": 0.5546875,
|
1275 |
+
"learning_rate": 5.5746561886051085e-06,
|
1276 |
+
"loss": 0.0246,
|
1277 |
+
"step": 1810
|
1278 |
+
},
|
1279 |
+
{
|
1280 |
+
"epoch": 0.8939096267190569,
|
1281 |
+
"grad_norm": 0.8671875,
|
1282 |
+
"learning_rate": 5.329076620825148e-06,
|
1283 |
+
"loss": 0.0249,
|
1284 |
+
"step": 1820
|
1285 |
+
},
|
1286 |
+
{
|
1287 |
+
"epoch": 0.8988212180746562,
|
1288 |
+
"grad_norm": 0.55859375,
|
1289 |
+
"learning_rate": 5.0834970530451866e-06,
|
1290 |
+
"loss": 0.0267,
|
1291 |
+
"step": 1830
|
1292 |
+
},
|
1293 |
+
{
|
1294 |
+
"epoch": 0.9037328094302554,
|
1295 |
+
"grad_norm": 0.5703125,
|
1296 |
+
"learning_rate": 4.837917485265226e-06,
|
1297 |
+
"loss": 0.0232,
|
1298 |
+
"step": 1840
|
1299 |
+
},
|
1300 |
+
{
|
1301 |
+
"epoch": 0.9086444007858546,
|
1302 |
+
"grad_norm": 0.6953125,
|
1303 |
+
"learning_rate": 4.5923379174852655e-06,
|
1304 |
+
"loss": 0.0242,
|
1305 |
+
"step": 1850
|
1306 |
+
},
|
1307 |
+
{
|
1308 |
+
"epoch": 0.9135559921414538,
|
1309 |
+
"grad_norm": 0.71484375,
|
1310 |
+
"learning_rate": 4.346758349705305e-06,
|
1311 |
+
"loss": 0.0256,
|
1312 |
+
"step": 1860
|
1313 |
+
},
|
1314 |
+
{
|
1315 |
+
"epoch": 0.918467583497053,
|
1316 |
+
"grad_norm": 1.125,
|
1317 |
+
"learning_rate": 4.1011787819253435e-06,
|
1318 |
+
"loss": 0.0318,
|
1319 |
+
"step": 1870
|
1320 |
+
},
|
1321 |
+
{
|
1322 |
+
"epoch": 0.9233791748526523,
|
1323 |
+
"grad_norm": 0.69921875,
|
1324 |
+
"learning_rate": 3.855599214145384e-06,
|
1325 |
+
"loss": 0.0268,
|
1326 |
+
"step": 1880
|
1327 |
+
},
|
1328 |
+
{
|
1329 |
+
"epoch": 0.9282907662082515,
|
1330 |
+
"grad_norm": 0.58203125,
|
1331 |
+
"learning_rate": 3.6100196463654228e-06,
|
1332 |
+
"loss": 0.0249,
|
1333 |
+
"step": 1890
|
1334 |
+
},
|
1335 |
+
{
|
1336 |
+
"epoch": 0.9332023575638507,
|
1337 |
+
"grad_norm": 0.66796875,
|
1338 |
+
"learning_rate": 3.364440078585462e-06,
|
1339 |
+
"loss": 0.0306,
|
1340 |
+
"step": 1900
|
1341 |
+
},
|
1342 |
+
{
|
1343 |
+
"epoch": 0.9381139489194499,
|
1344 |
+
"grad_norm": 0.59765625,
|
1345 |
+
"learning_rate": 3.1188605108055012e-06,
|
1346 |
+
"loss": 0.0301,
|
1347 |
+
"step": 1910
|
1348 |
+
},
|
1349 |
+
{
|
1350 |
+
"epoch": 0.9430255402750491,
|
1351 |
+
"grad_norm": 0.65625,
|
1352 |
+
"learning_rate": 2.8732809430255403e-06,
|
1353 |
+
"loss": 0.0248,
|
1354 |
+
"step": 1920
|
1355 |
+
},
|
1356 |
+
{
|
1357 |
+
"epoch": 0.9479371316306483,
|
1358 |
+
"grad_norm": 0.76953125,
|
1359 |
+
"learning_rate": 2.6277013752455797e-06,
|
1360 |
+
"loss": 0.0267,
|
1361 |
+
"step": 1930
|
1362 |
+
},
|
1363 |
+
{
|
1364 |
+
"epoch": 0.9528487229862476,
|
1365 |
+
"grad_norm": 0.61328125,
|
1366 |
+
"learning_rate": 2.382121807465619e-06,
|
1367 |
+
"loss": 0.0296,
|
1368 |
+
"step": 1940
|
1369 |
+
},
|
1370 |
+
{
|
1371 |
+
"epoch": 0.9577603143418467,
|
1372 |
+
"grad_norm": 0.59375,
|
1373 |
+
"learning_rate": 2.136542239685658e-06,
|
1374 |
+
"loss": 0.0264,
|
1375 |
+
"step": 1950
|
1376 |
+
},
|
1377 |
+
{
|
1378 |
+
"epoch": 0.962671905697446,
|
1379 |
+
"grad_norm": 0.51953125,
|
1380 |
+
"learning_rate": 1.8909626719056976e-06,
|
1381 |
+
"loss": 0.024,
|
1382 |
+
"step": 1960
|
1383 |
+
},
|
1384 |
+
{
|
1385 |
+
"epoch": 0.9675834970530451,
|
1386 |
+
"grad_norm": 0.56640625,
|
1387 |
+
"learning_rate": 1.6453831041257368e-06,
|
1388 |
+
"loss": 0.0315,
|
1389 |
+
"step": 1970
|
1390 |
+
},
|
1391 |
+
{
|
1392 |
+
"epoch": 0.9724950884086444,
|
1393 |
+
"grad_norm": 0.859375,
|
1394 |
+
"learning_rate": 1.3998035363457763e-06,
|
1395 |
+
"loss": 0.024,
|
1396 |
+
"step": 1980
|
1397 |
+
},
|
1398 |
+
{
|
1399 |
+
"epoch": 0.9774066797642437,
|
1400 |
+
"grad_norm": 0.8125,
|
1401 |
+
"learning_rate": 1.1542239685658153e-06,
|
1402 |
+
"loss": 0.0282,
|
1403 |
+
"step": 1990
|
1404 |
+
},
|
1405 |
+
{
|
1406 |
+
"epoch": 0.9823182711198428,
|
1407 |
+
"grad_norm": 0.6328125,
|
1408 |
+
"learning_rate": 9.086444007858547e-07,
|
1409 |
+
"loss": 0.0231,
|
1410 |
+
"step": 2000
|
1411 |
+
},
|
1412 |
+
{
|
1413 |
+
"epoch": 0.9872298624754421,
|
1414 |
+
"grad_norm": 0.75390625,
|
1415 |
+
"learning_rate": 6.630648330058939e-07,
|
1416 |
+
"loss": 0.0247,
|
1417 |
+
"step": 2010
|
1418 |
+
},
|
1419 |
+
{
|
1420 |
+
"epoch": 0.9921414538310412,
|
1421 |
+
"grad_norm": 0.5546875,
|
1422 |
+
"learning_rate": 4.174852652259332e-07,
|
1423 |
+
"loss": 0.0233,
|
1424 |
+
"step": 2020
|
1425 |
+
},
|
1426 |
+
{
|
1427 |
+
"epoch": 0.9970530451866405,
|
1428 |
+
"grad_norm": 0.85546875,
|
1429 |
+
"learning_rate": 1.719056974459725e-07,
|
1430 |
+
"loss": 0.0366,
|
1431 |
+
"step": 2030
|
1432 |
+
}
|
1433 |
+
],
|
1434 |
+
"logging_steps": 10,
|
1435 |
+
"max_steps": 2036,
|
1436 |
+
"num_input_tokens_seen": 0,
|
1437 |
+
"num_train_epochs": 1,
|
1438 |
+
"save_steps": 500,
|
1439 |
+
"stateful_callbacks": {
|
1440 |
+
"TrainerControl": {
|
1441 |
+
"args": {
|
1442 |
+
"should_epoch_stop": false,
|
1443 |
+
"should_evaluate": false,
|
1444 |
+
"should_log": false,
|
1445 |
+
"should_save": true,
|
1446 |
+
"should_training_stop": true
|
1447 |
+
},
|
1448 |
+
"attributes": {}
|
1449 |
+
}
|
1450 |
+
},
|
1451 |
+
"total_flos": 1.9084808495824896e+16,
|
1452 |
+
"train_batch_size": 2,
|
1453 |
+
"trial_name": null,
|
1454 |
+
"trial_params": null
|
1455 |
+
}
|
last-checkpoint/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:96b4692fcf47c9552185532e3f4edfd1a00198142146dd822a7aa35f5d117280
|
3 |
+
size 5304
|