Upload 14 files
Browse files- added_tokens.json +9 -0
- config.json +38 -0
- generation_config.json +6 -0
- merges.txt +0 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scaler.pt +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +53 -0
- tokenizer_config.json +116 -0
- trainer_state.json +1934 -0
- training_args.bin +3 -0
- vocab.json +0 -0
added_tokens.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<|câu6|>": 50261,
|
| 3 |
+
"<|câu8|>": 50262,
|
| 4 |
+
"<|endoftext|>": 50257,
|
| 5 |
+
"<|khổ|>": 50263,
|
| 6 |
+
"<|kết|>": 50259,
|
| 7 |
+
"<|pad|>": 50258,
|
| 8 |
+
"<|thơ|>": 50260
|
| 9 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,38 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_function": "gelu_new",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"GPT2LMHeadModel"
|
| 5 |
+
],
|
| 6 |
+
"attn_pdrop": 0.0,
|
| 7 |
+
"bos_token_id": 50256,
|
| 8 |
+
"embd_pdrop": 0.0,
|
| 9 |
+
"eos_token_id": 50256,
|
| 10 |
+
"initializer_range": 0.02,
|
| 11 |
+
"layer_norm_epsilon": 1e-05,
|
| 12 |
+
"model_type": "gpt2",
|
| 13 |
+
"n_ctx": 1024,
|
| 14 |
+
"n_embd": 768,
|
| 15 |
+
"n_head": 12,
|
| 16 |
+
"n_inner": null,
|
| 17 |
+
"n_layer": 12,
|
| 18 |
+
"n_positions": 1024,
|
| 19 |
+
"reorder_and_upcast_attn": false,
|
| 20 |
+
"resid_pdrop": 0.0,
|
| 21 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 22 |
+
"scale_attn_weights": true,
|
| 23 |
+
"summary_activation": null,
|
| 24 |
+
"summary_first_dropout": 0.1,
|
| 25 |
+
"summary_proj_to_labels": true,
|
| 26 |
+
"summary_type": "cls_index",
|
| 27 |
+
"summary_use_proj": true,
|
| 28 |
+
"task_specific_params": {
|
| 29 |
+
"text-generation": {
|
| 30 |
+
"do_sample": true,
|
| 31 |
+
"max_length": 50
|
| 32 |
+
}
|
| 33 |
+
},
|
| 34 |
+
"torch_dtype": "float32",
|
| 35 |
+
"transformers_version": "4.51.3",
|
| 36 |
+
"use_cache": true,
|
| 37 |
+
"vocab_size": 50264
|
| 38 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 50256,
|
| 4 |
+
"eos_token_id": 50256,
|
| 5 |
+
"transformers_version": "4.51.3"
|
| 6 |
+
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:68605f8a459cefde6fa7afa47ed6de1be5bb816422175c8aeecb53c816457c3f
|
| 3 |
+
size 497795712
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:652c2b0dca5a9b2b71c9428329eb11e91cbe7a52b6df58b5d95d456a4cd67b88
|
| 3 |
+
size 995685771
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f4b94d24c5fd3f8add3c432b54f5305b9646b5d564b95bdc0317d59c71a8bc1c
|
| 3 |
+
size 14645
|
scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d7d1729a0a737781a04316ceb381d8571ac00a4cfb9935801904a3a63d931dc3
|
| 3 |
+
size 1383
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:45c7433081939794c4f3f9ca0a51bae55a2d89749e5802ef10d8b5cec4de5547
|
| 3 |
+
size 1465
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,53 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
{
|
| 4 |
+
"content": "<|câu6|>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false
|
| 9 |
+
},
|
| 10 |
+
{
|
| 11 |
+
"content": "<|câu8|>",
|
| 12 |
+
"lstrip": false,
|
| 13 |
+
"normalized": false,
|
| 14 |
+
"rstrip": false,
|
| 15 |
+
"single_word": false
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"content": "<|khổ|>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": false,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
}
|
| 24 |
+
],
|
| 25 |
+
"bos_token": {
|
| 26 |
+
"content": "<|thơ|>",
|
| 27 |
+
"lstrip": false,
|
| 28 |
+
"normalized": false,
|
| 29 |
+
"rstrip": false,
|
| 30 |
+
"single_word": false
|
| 31 |
+
},
|
| 32 |
+
"eos_token": {
|
| 33 |
+
"content": "<|kết|>",
|
| 34 |
+
"lstrip": false,
|
| 35 |
+
"normalized": false,
|
| 36 |
+
"rstrip": false,
|
| 37 |
+
"single_word": false
|
| 38 |
+
},
|
| 39 |
+
"pad_token": {
|
| 40 |
+
"content": "<|pad|>",
|
| 41 |
+
"lstrip": false,
|
| 42 |
+
"normalized": false,
|
| 43 |
+
"rstrip": false,
|
| 44 |
+
"single_word": false
|
| 45 |
+
},
|
| 46 |
+
"unk_token": {
|
| 47 |
+
"content": "<|endoftext|>",
|
| 48 |
+
"lstrip": false,
|
| 49 |
+
"normalized": false,
|
| 50 |
+
"rstrip": false,
|
| 51 |
+
"single_word": false
|
| 52 |
+
}
|
| 53 |
+
}
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,116 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": false,
|
| 3 |
+
"add_prefix_space": false,
|
| 4 |
+
"added_tokens_decoder": {
|
| 5 |
+
"0": {
|
| 6 |
+
"content": "<s>",
|
| 7 |
+
"lstrip": false,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false,
|
| 11 |
+
"special": true
|
| 12 |
+
},
|
| 13 |
+
"1": {
|
| 14 |
+
"content": "<pad>",
|
| 15 |
+
"lstrip": false,
|
| 16 |
+
"normalized": false,
|
| 17 |
+
"rstrip": false,
|
| 18 |
+
"single_word": false,
|
| 19 |
+
"special": true
|
| 20 |
+
},
|
| 21 |
+
"2": {
|
| 22 |
+
"content": "</s>",
|
| 23 |
+
"lstrip": false,
|
| 24 |
+
"normalized": false,
|
| 25 |
+
"rstrip": false,
|
| 26 |
+
"single_word": false,
|
| 27 |
+
"special": true
|
| 28 |
+
},
|
| 29 |
+
"3": {
|
| 30 |
+
"content": "<unk>",
|
| 31 |
+
"lstrip": false,
|
| 32 |
+
"normalized": false,
|
| 33 |
+
"rstrip": false,
|
| 34 |
+
"single_word": false,
|
| 35 |
+
"special": true
|
| 36 |
+
},
|
| 37 |
+
"4": {
|
| 38 |
+
"content": "<mask>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": false,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false,
|
| 43 |
+
"special": true
|
| 44 |
+
},
|
| 45 |
+
"50257": {
|
| 46 |
+
"content": "<|endoftext|>",
|
| 47 |
+
"lstrip": false,
|
| 48 |
+
"normalized": false,
|
| 49 |
+
"rstrip": false,
|
| 50 |
+
"single_word": false,
|
| 51 |
+
"special": true
|
| 52 |
+
},
|
| 53 |
+
"50258": {
|
| 54 |
+
"content": "<|pad|>",
|
| 55 |
+
"lstrip": false,
|
| 56 |
+
"normalized": false,
|
| 57 |
+
"rstrip": false,
|
| 58 |
+
"single_word": false,
|
| 59 |
+
"special": true
|
| 60 |
+
},
|
| 61 |
+
"50259": {
|
| 62 |
+
"content": "<|kết|>",
|
| 63 |
+
"lstrip": false,
|
| 64 |
+
"normalized": false,
|
| 65 |
+
"rstrip": false,
|
| 66 |
+
"single_word": false,
|
| 67 |
+
"special": true
|
| 68 |
+
},
|
| 69 |
+
"50260": {
|
| 70 |
+
"content": "<|thơ|>",
|
| 71 |
+
"lstrip": false,
|
| 72 |
+
"normalized": false,
|
| 73 |
+
"rstrip": false,
|
| 74 |
+
"single_word": false,
|
| 75 |
+
"special": true
|
| 76 |
+
},
|
| 77 |
+
"50261": {
|
| 78 |
+
"content": "<|câu6|>",
|
| 79 |
+
"lstrip": false,
|
| 80 |
+
"normalized": false,
|
| 81 |
+
"rstrip": false,
|
| 82 |
+
"single_word": false,
|
| 83 |
+
"special": true
|
| 84 |
+
},
|
| 85 |
+
"50262": {
|
| 86 |
+
"content": "<|câu8|>",
|
| 87 |
+
"lstrip": false,
|
| 88 |
+
"normalized": false,
|
| 89 |
+
"rstrip": false,
|
| 90 |
+
"single_word": false,
|
| 91 |
+
"special": true
|
| 92 |
+
},
|
| 93 |
+
"50263": {
|
| 94 |
+
"content": "<|khổ|>",
|
| 95 |
+
"lstrip": false,
|
| 96 |
+
"normalized": false,
|
| 97 |
+
"rstrip": false,
|
| 98 |
+
"single_word": false,
|
| 99 |
+
"special": true
|
| 100 |
+
}
|
| 101 |
+
},
|
| 102 |
+
"additional_special_tokens": [
|
| 103 |
+
"<|câu6|>",
|
| 104 |
+
"<|câu8|>",
|
| 105 |
+
"<|khổ|>"
|
| 106 |
+
],
|
| 107 |
+
"bos_token": "<|thơ|>",
|
| 108 |
+
"clean_up_tokenization_spaces": false,
|
| 109 |
+
"eos_token": "<|kết|>",
|
| 110 |
+
"errors": "replace",
|
| 111 |
+
"extra_special_tokens": {},
|
| 112 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 113 |
+
"pad_token": "<|pad|>",
|
| 114 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 115 |
+
"unk_token": "<|endoftext|>"
|
| 116 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1934 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 6681,
|
| 3 |
+
"best_metric": 3.168698787689209,
|
| 4 |
+
"best_model_checkpoint": "./luc-bat-poet-model\\checkpoint-6681",
|
| 5 |
+
"epoch": 3.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 6681,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.00044917324050419694,
|
| 14 |
+
"grad_norm": 5.473126411437988,
|
| 15 |
+
"learning_rate": 0.0,
|
| 16 |
+
"loss": 5.4778,
|
| 17 |
+
"step": 1
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.011229331012604924,
|
| 21 |
+
"grad_norm": 0.8930467963218689,
|
| 22 |
+
"learning_rate": 0.00012,
|
| 23 |
+
"loss": 5.0452,
|
| 24 |
+
"step": 25
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.02245866202520985,
|
| 28 |
+
"grad_norm": 0.7617964744567871,
|
| 29 |
+
"learning_rate": 0.000245,
|
| 30 |
+
"loss": 4.5144,
|
| 31 |
+
"step": 50
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.03368799303781477,
|
| 35 |
+
"grad_norm": 0.8281979560852051,
|
| 36 |
+
"learning_rate": 0.00037,
|
| 37 |
+
"loss": 4.3613,
|
| 38 |
+
"step": 75
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.0449173240504197,
|
| 42 |
+
"grad_norm": 0.8189941644668579,
|
| 43 |
+
"learning_rate": 0.000495,
|
| 44 |
+
"loss": 4.2239,
|
| 45 |
+
"step": 100
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.05614665506302462,
|
| 49 |
+
"grad_norm": 1.01946222782135,
|
| 50 |
+
"learning_rate": 0.00062,
|
| 51 |
+
"loss": 4.1386,
|
| 52 |
+
"step": 125
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.06737598607562954,
|
| 56 |
+
"grad_norm": 0.9489204287528992,
|
| 57 |
+
"learning_rate": 0.000745,
|
| 58 |
+
"loss": 4.0065,
|
| 59 |
+
"step": 150
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.07860531708823447,
|
| 63 |
+
"grad_norm": 1.6589257717132568,
|
| 64 |
+
"learning_rate": 0.00087,
|
| 65 |
+
"loss": 3.9373,
|
| 66 |
+
"step": 175
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.0898346481008394,
|
| 70 |
+
"grad_norm": 1.1351829767227173,
|
| 71 |
+
"learning_rate": 0.000995,
|
| 72 |
+
"loss": 3.9028,
|
| 73 |
+
"step": 200
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.10106397911344432,
|
| 77 |
+
"grad_norm": 0.7443220615386963,
|
| 78 |
+
"learning_rate": 0.000999995416032659,
|
| 79 |
+
"loss": 3.8687,
|
| 80 |
+
"step": 225
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.11229331012604923,
|
| 84 |
+
"grad_norm": 0.8907812833786011,
|
| 85 |
+
"learning_rate": 0.0009999808922703088,
|
| 86 |
+
"loss": 3.8144,
|
| 87 |
+
"step": 250
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.12352264113865416,
|
| 91 |
+
"grad_norm": 0.651372492313385,
|
| 92 |
+
"learning_rate": 0.0009999564210436207,
|
| 93 |
+
"loss": 3.8007,
|
| 94 |
+
"step": 275
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.1347519721512591,
|
| 98 |
+
"grad_norm": 0.7908700704574585,
|
| 99 |
+
"learning_rate": 0.000999922002839467,
|
| 100 |
+
"loss": 3.7684,
|
| 101 |
+
"step": 300
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.145981303163864,
|
| 105 |
+
"grad_norm": 0.6001281142234802,
|
| 106 |
+
"learning_rate": 0.0009998776383426215,
|
| 107 |
+
"loss": 3.743,
|
| 108 |
+
"step": 325
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.15721063417646894,
|
| 112 |
+
"grad_norm": 0.6484293937683105,
|
| 113 |
+
"learning_rate": 0.0009998233284357462,
|
| 114 |
+
"loss": 3.7416,
|
| 115 |
+
"step": 350
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.16843996518907386,
|
| 119 |
+
"grad_norm": 0.7714053988456726,
|
| 120 |
+
"learning_rate": 0.0009997590741993743,
|
| 121 |
+
"loss": 3.7108,
|
| 122 |
+
"step": 375
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.1796692962016788,
|
| 126 |
+
"grad_norm": 0.6280301213264465,
|
| 127 |
+
"learning_rate": 0.0009996848769118882,
|
| 128 |
+
"loss": 3.6854,
|
| 129 |
+
"step": 400
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.19089862721428372,
|
| 133 |
+
"grad_norm": 0.644794225692749,
|
| 134 |
+
"learning_rate": 0.0009996007380494937,
|
| 135 |
+
"loss": 3.6737,
|
| 136 |
+
"step": 425
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.20212795822688864,
|
| 140 |
+
"grad_norm": 0.6872850060462952,
|
| 141 |
+
"learning_rate": 0.0009995066592861919,
|
| 142 |
+
"loss": 3.662,
|
| 143 |
+
"step": 450
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.21335728923949357,
|
| 147 |
+
"grad_norm": 0.601828396320343,
|
| 148 |
+
"learning_rate": 0.0009994026424937441,
|
| 149 |
+
"loss": 3.678,
|
| 150 |
+
"step": 475
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.22458662025209847,
|
| 154 |
+
"grad_norm": 0.6280804872512817,
|
| 155 |
+
"learning_rate": 0.0009992886897416365,
|
| 156 |
+
"loss": 3.6652,
|
| 157 |
+
"step": 500
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.2358159512647034,
|
| 161 |
+
"grad_norm": 0.5711939930915833,
|
| 162 |
+
"learning_rate": 0.0009991648032970373,
|
| 163 |
+
"loss": 3.627,
|
| 164 |
+
"step": 525
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.24704528227730832,
|
| 168 |
+
"grad_norm": 0.5883836150169373,
|
| 169 |
+
"learning_rate": 0.000999030985624753,
|
| 170 |
+
"loss": 3.6433,
|
| 171 |
+
"step": 550
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.25827461328991325,
|
| 175 |
+
"grad_norm": 0.5956864356994629,
|
| 176 |
+
"learning_rate": 0.000998887239387178,
|
| 177 |
+
"loss": 3.6207,
|
| 178 |
+
"step": 575
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.2695039443025182,
|
| 182 |
+
"grad_norm": 0.577458918094635,
|
| 183 |
+
"learning_rate": 0.000998733567444243,
|
| 184 |
+
"loss": 3.6074,
|
| 185 |
+
"step": 600
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.2807332753151231,
|
| 189 |
+
"grad_norm": 0.8079866170883179,
|
| 190 |
+
"learning_rate": 0.0009985699728533573,
|
| 191 |
+
"loss": 3.6119,
|
| 192 |
+
"step": 625
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.291962606327728,
|
| 196 |
+
"grad_norm": 0.5645362138748169,
|
| 197 |
+
"learning_rate": 0.0009983964588693478,
|
| 198 |
+
"loss": 3.6411,
|
| 199 |
+
"step": 650
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.30319193734033295,
|
| 203 |
+
"grad_norm": 0.5666602253913879,
|
| 204 |
+
"learning_rate": 0.0009982130289443944,
|
| 205 |
+
"loss": 3.5725,
|
| 206 |
+
"step": 675
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.3144212683529379,
|
| 210 |
+
"grad_norm": 0.6160576343536377,
|
| 211 |
+
"learning_rate": 0.0009980196867279626,
|
| 212 |
+
"loss": 3.6055,
|
| 213 |
+
"step": 700
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.3256505993655428,
|
| 217 |
+
"grad_norm": 0.6656046509742737,
|
| 218 |
+
"learning_rate": 0.0009978164360667286,
|
| 219 |
+
"loss": 3.5776,
|
| 220 |
+
"step": 725
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.33687993037814773,
|
| 224 |
+
"grad_norm": 0.6236292123794556,
|
| 225 |
+
"learning_rate": 0.0009976032810045043,
|
| 226 |
+
"loss": 3.5509,
|
| 227 |
+
"step": 750
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.34810926139075266,
|
| 231 |
+
"grad_norm": 0.5736802816390991,
|
| 232 |
+
"learning_rate": 0.0009973802257821566,
|
| 233 |
+
"loss": 3.557,
|
| 234 |
+
"step": 775
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.3593385924033576,
|
| 238 |
+
"grad_norm": 0.5729912519454956,
|
| 239 |
+
"learning_rate": 0.000997147274837523,
|
| 240 |
+
"loss": 3.5374,
|
| 241 |
+
"step": 800
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.3705679234159625,
|
| 245 |
+
"grad_norm": 0.6009777188301086,
|
| 246 |
+
"learning_rate": 0.000996904432805323,
|
| 247 |
+
"loss": 3.5564,
|
| 248 |
+
"step": 825
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.38179725442856743,
|
| 252 |
+
"grad_norm": 0.5442889928817749,
|
| 253 |
+
"learning_rate": 0.0009966517045170659,
|
| 254 |
+
"loss": 3.5109,
|
| 255 |
+
"step": 850
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.39302658544117236,
|
| 259 |
+
"grad_norm": 0.5199196338653564,
|
| 260 |
+
"learning_rate": 0.0009963890950009549,
|
| 261 |
+
"loss": 3.526,
|
| 262 |
+
"step": 875
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.4042559164537773,
|
| 266 |
+
"grad_norm": 0.5444336533546448,
|
| 267 |
+
"learning_rate": 0.000996116609481788,
|
| 268 |
+
"loss": 3.5177,
|
| 269 |
+
"step": 900
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.4154852474663822,
|
| 273 |
+
"grad_norm": 0.5710541009902954,
|
| 274 |
+
"learning_rate": 0.000995834253380852,
|
| 275 |
+
"loss": 3.4767,
|
| 276 |
+
"step": 925
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.42671457847898714,
|
| 280 |
+
"grad_norm": 0.5203866362571716,
|
| 281 |
+
"learning_rate": 0.000995542032315816,
|
| 282 |
+
"loss": 3.4972,
|
| 283 |
+
"step": 950
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.43794390949159206,
|
| 287 |
+
"grad_norm": 0.545213520526886,
|
| 288 |
+
"learning_rate": 0.0009952399521006192,
|
| 289 |
+
"loss": 3.4623,
|
| 290 |
+
"step": 975
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.44917324050419694,
|
| 294 |
+
"grad_norm": 0.5384295582771301,
|
| 295 |
+
"learning_rate": 0.0009949280187453561,
|
| 296 |
+
"loss": 3.4367,
|
| 297 |
+
"step": 1000
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.46040257151680186,
|
| 301 |
+
"grad_norm": 0.5814492106437683,
|
| 302 |
+
"learning_rate": 0.0009946062384561555,
|
| 303 |
+
"loss": 3.441,
|
| 304 |
+
"step": 1025
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.4716319025294068,
|
| 308 |
+
"grad_norm": 0.5313106179237366,
|
| 309 |
+
"learning_rate": 0.000994274617635058,
|
| 310 |
+
"loss": 3.4336,
|
| 311 |
+
"step": 1050
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.4828612335420117,
|
| 315 |
+
"grad_norm": 0.6174953579902649,
|
| 316 |
+
"learning_rate": 0.0009939331628798882,
|
| 317 |
+
"loss": 3.3919,
|
| 318 |
+
"step": 1075
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.49409056455461664,
|
| 322 |
+
"grad_norm": 0.566247284412384,
|
| 323 |
+
"learning_rate": 0.0009935818809841239,
|
| 324 |
+
"loss": 3.4281,
|
| 325 |
+
"step": 1100
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.5053198955672216,
|
| 329 |
+
"grad_norm": 0.5577934384346008,
|
| 330 |
+
"learning_rate": 0.0009932207789367603,
|
| 331 |
+
"loss": 3.4043,
|
| 332 |
+
"step": 1125
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.5165492265798265,
|
| 336 |
+
"grad_norm": 0.5082374811172485,
|
| 337 |
+
"learning_rate": 0.0009928498639221715,
|
| 338 |
+
"loss": 3.3853,
|
| 339 |
+
"step": 1150
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.5277785575924314,
|
| 343 |
+
"grad_norm": 0.49451902508735657,
|
| 344 |
+
"learning_rate": 0.0009924691433199674,
|
| 345 |
+
"loss": 3.3794,
|
| 346 |
+
"step": 1175
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.5390078886050363,
|
| 350 |
+
"grad_norm": 0.5019585490226746,
|
| 351 |
+
"learning_rate": 0.0009920786247048464,
|
| 352 |
+
"loss": 3.4127,
|
| 353 |
+
"step": 1200
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.5502372196176413,
|
| 357 |
+
"grad_norm": 0.599949300289154,
|
| 358 |
+
"learning_rate": 0.0009916783158464455,
|
| 359 |
+
"loss": 3.3923,
|
| 360 |
+
"step": 1225
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.5614665506302462,
|
| 364 |
+
"grad_norm": 0.5792480707168579,
|
| 365 |
+
"learning_rate": 0.0009912682247091853,
|
| 366 |
+
"loss": 3.3656,
|
| 367 |
+
"step": 1250
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.5726958816428511,
|
| 371 |
+
"grad_norm": 0.5839366912841797,
|
| 372 |
+
"learning_rate": 0.0009908483594521116,
|
| 373 |
+
"loss": 3.3895,
|
| 374 |
+
"step": 1275
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.583925212655456,
|
| 378 |
+
"grad_norm": 0.5739409923553467,
|
| 379 |
+
"learning_rate": 0.0009904187284287332,
|
| 380 |
+
"loss": 3.3506,
|
| 381 |
+
"step": 1300
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.595154543668061,
|
| 385 |
+
"grad_norm": 0.5897430777549744,
|
| 386 |
+
"learning_rate": 0.0009899793401868546,
|
| 387 |
+
"loss": 3.3247,
|
| 388 |
+
"step": 1325
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.6063838746806659,
|
| 392 |
+
"grad_norm": 0.5063382983207703,
|
| 393 |
+
"learning_rate": 0.0009895302034684083,
|
| 394 |
+
"loss": 3.3369,
|
| 395 |
+
"step": 1350
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.6176132056932708,
|
| 399 |
+
"grad_norm": 0.5431721806526184,
|
| 400 |
+
"learning_rate": 0.0009890713272092786,
|
| 401 |
+
"loss": 3.3446,
|
| 402 |
+
"step": 1375
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.6288425367058758,
|
| 406 |
+
"grad_norm": 0.4924439489841461,
|
| 407 |
+
"learning_rate": 0.0009886027205391248,
|
| 408 |
+
"loss": 3.3377,
|
| 409 |
+
"step": 1400
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.6400718677184807,
|
| 413 |
+
"grad_norm": 0.54843670129776,
|
| 414 |
+
"learning_rate": 0.0009881243927811992,
|
| 415 |
+
"loss": 3.345,
|
| 416 |
+
"step": 1425
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.6513011987310856,
|
| 420 |
+
"grad_norm": 0.5399184226989746,
|
| 421 |
+
"learning_rate": 0.0009876363534521626,
|
| 422 |
+
"loss": 3.3344,
|
| 423 |
+
"step": 1450
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.6625305297436905,
|
| 427 |
+
"grad_norm": 0.5677058696746826,
|
| 428 |
+
"learning_rate": 0.0009871386122618933,
|
| 429 |
+
"loss": 3.3328,
|
| 430 |
+
"step": 1475
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.6737598607562955,
|
| 434 |
+
"grad_norm": 0.482010155916214,
|
| 435 |
+
"learning_rate": 0.0009866311791132953,
|
| 436 |
+
"loss": 3.3433,
|
| 437 |
+
"step": 1500
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.6849891917689004,
|
| 441 |
+
"grad_norm": 0.49796995520591736,
|
| 442 |
+
"learning_rate": 0.000986114064102101,
|
| 443 |
+
"loss": 3.3413,
|
| 444 |
+
"step": 1525
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.6962185227815053,
|
| 448 |
+
"grad_norm": 0.4976087808609009,
|
| 449 |
+
"learning_rate": 0.0009855872775166696,
|
| 450 |
+
"loss": 3.3176,
|
| 451 |
+
"step": 1550
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.7074478537941102,
|
| 455 |
+
"grad_norm": 0.5586963891983032,
|
| 456 |
+
"learning_rate": 0.0009850508298377832,
|
| 457 |
+
"loss": 3.3188,
|
| 458 |
+
"step": 1575
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.7186771848067152,
|
| 462 |
+
"grad_norm": 0.45419466495513916,
|
| 463 |
+
"learning_rate": 0.0009845047317384378,
|
| 464 |
+
"loss": 3.2902,
|
| 465 |
+
"step": 1600
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.7299065158193201,
|
| 469 |
+
"grad_norm": 0.5402281880378723,
|
| 470 |
+
"learning_rate": 0.0009839489940836317,
|
| 471 |
+
"loss": 3.2893,
|
| 472 |
+
"step": 1625
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.741135846831925,
|
| 476 |
+
"grad_norm": 0.5439639687538147,
|
| 477 |
+
"learning_rate": 0.0009833836279301484,
|
| 478 |
+
"loss": 3.3065,
|
| 479 |
+
"step": 1650
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.7523651778445299,
|
| 483 |
+
"grad_norm": 0.49050387740135193,
|
| 484 |
+
"learning_rate": 0.0009828086445263368,
|
| 485 |
+
"loss": 3.2796,
|
| 486 |
+
"step": 1675
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.7635945088571349,
|
| 490 |
+
"grad_norm": 0.5266304016113281,
|
| 491 |
+
"learning_rate": 0.000982224055311888,
|
| 492 |
+
"loss": 3.29,
|
| 493 |
+
"step": 1700
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.7748238398697398,
|
| 497 |
+
"grad_norm": 0.5335237979888916,
|
| 498 |
+
"learning_rate": 0.0009816298719176073,
|
| 499 |
+
"loss": 3.2889,
|
| 500 |
+
"step": 1725
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.7860531708823447,
|
| 504 |
+
"grad_norm": 0.5236843824386597,
|
| 505 |
+
"learning_rate": 0.0009810261061651826,
|
| 506 |
+
"loss": 3.2758,
|
| 507 |
+
"step": 1750
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.7972825018949496,
|
| 511 |
+
"grad_norm": 0.541266679763794,
|
| 512 |
+
"learning_rate": 0.0009804127700669496,
|
| 513 |
+
"loss": 3.3053,
|
| 514 |
+
"step": 1775
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.8085118329075546,
|
| 518 |
+
"grad_norm": 0.47298872470855713,
|
| 519 |
+
"learning_rate": 0.0009797898758256525,
|
| 520 |
+
"loss": 3.277,
|
| 521 |
+
"step": 1800
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.8197411639201595,
|
| 525 |
+
"grad_norm": 0.5097940564155579,
|
| 526 |
+
"learning_rate": 0.0009791574358342014,
|
| 527 |
+
"loss": 3.2537,
|
| 528 |
+
"step": 1825
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.8309704949327644,
|
| 532 |
+
"grad_norm": 0.5668537616729736,
|
| 533 |
+
"learning_rate": 0.0009785154626754259,
|
| 534 |
+
"loss": 3.2514,
|
| 535 |
+
"step": 1850
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.8421998259453694,
|
| 539 |
+
"grad_norm": 0.5071256160736084,
|
| 540 |
+
"learning_rate": 0.000977863969121824,
|
| 541 |
+
"loss": 3.2467,
|
| 542 |
+
"step": 1875
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.8534291569579743,
|
| 546 |
+
"grad_norm": 0.4892118573188782,
|
| 547 |
+
"learning_rate": 0.000977202968135309,
|
| 548 |
+
"loss": 3.2772,
|
| 549 |
+
"step": 1900
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.8646584879705792,
|
| 553 |
+
"grad_norm": 0.463008850812912,
|
| 554 |
+
"learning_rate": 0.000976532472866951,
|
| 555 |
+
"loss": 3.2492,
|
| 556 |
+
"step": 1925
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.8758878189831841,
|
| 560 |
+
"grad_norm": 0.5179749727249146,
|
| 561 |
+
"learning_rate": 0.0009758524966567152,
|
| 562 |
+
"loss": 3.2289,
|
| 563 |
+
"step": 1950
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.8871171499957891,
|
| 567 |
+
"grad_norm": 0.5276876091957092,
|
| 568 |
+
"learning_rate": 0.000975163053033197,
|
| 569 |
+
"loss": 3.2474,
|
| 570 |
+
"step": 1975
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.8983464810083939,
|
| 574 |
+
"grad_norm": 0.5618298649787903,
|
| 575 |
+
"learning_rate": 0.000974464155713352,
|
| 576 |
+
"loss": 3.2652,
|
| 577 |
+
"step": 2000
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.9095758120209988,
|
| 581 |
+
"grad_norm": 0.4800003468990326,
|
| 582 |
+
"learning_rate": 0.0009737558186022242,
|
| 583 |
+
"loss": 3.2424,
|
| 584 |
+
"step": 2025
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.9208051430336037,
|
| 588 |
+
"grad_norm": 0.47998154163360596,
|
| 589 |
+
"learning_rate": 0.0009730380557926682,
|
| 590 |
+
"loss": 3.2737,
|
| 591 |
+
"step": 2050
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.9320344740462086,
|
| 595 |
+
"grad_norm": 0.5447694659233093,
|
| 596 |
+
"learning_rate": 0.00097231088156507,
|
| 597 |
+
"loss": 3.249,
|
| 598 |
+
"step": 2075
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.9432638050588136,
|
| 602 |
+
"grad_norm": 0.5559055209159851,
|
| 603 |
+
"learning_rate": 0.0009715743103870615,
|
| 604 |
+
"loss": 3.2566,
|
| 605 |
+
"step": 2100
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.9544931360714185,
|
| 609 |
+
"grad_norm": 0.478614866733551,
|
| 610 |
+
"learning_rate": 0.0009708283569132341,
|
| 611 |
+
"loss": 3.2076,
|
| 612 |
+
"step": 2125
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.9657224670840234,
|
| 616 |
+
"grad_norm": 0.44457143545150757,
|
| 617 |
+
"learning_rate": 0.000970073035984846,
|
| 618 |
+
"loss": 3.2052,
|
| 619 |
+
"step": 2150
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.9769517980966284,
|
| 623 |
+
"grad_norm": 0.5057160258293152,
|
| 624 |
+
"learning_rate": 0.0009693083626295274,
|
| 625 |
+
"loss": 3.1944,
|
| 626 |
+
"step": 2175
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.9881811291092333,
|
| 630 |
+
"grad_norm": 0.487543523311615,
|
| 631 |
+
"learning_rate": 0.0009685343520609816,
|
| 632 |
+
"loss": 3.2862,
|
| 633 |
+
"step": 2200
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.9994104601218382,
|
| 637 |
+
"grad_norm": 0.5547086000442505,
|
| 638 |
+
"learning_rate": 0.0009677510196786822,
|
| 639 |
+
"loss": 3.2249,
|
| 640 |
+
"step": 2225
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 1.0,
|
| 644 |
+
"eval_loss": 3.2925968170166016,
|
| 645 |
+
"eval_runtime": 230.4828,
|
| 646 |
+
"eval_samples_per_second": 54.598,
|
| 647 |
+
"eval_steps_per_second": 54.598,
|
| 648 |
+
"step": 2227
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 1.0103309845315964,
|
| 652 |
+
"grad_norm": 0.5509684085845947,
|
| 653 |
+
"learning_rate": 0.0009669583810675666,
|
| 654 |
+
"loss": 3.0297,
|
| 655 |
+
"step": 2250
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 1.0215603155442015,
|
| 659 |
+
"grad_norm": 0.5036989450454712,
|
| 660 |
+
"learning_rate": 0.0009661564519977263,
|
| 661 |
+
"loss": 2.9815,
|
| 662 |
+
"step": 2275
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 1.0327896465568063,
|
| 666 |
+
"grad_norm": 0.5602796673774719,
|
| 667 |
+
"learning_rate": 0.0009653452484240923,
|
| 668 |
+
"loss": 2.994,
|
| 669 |
+
"step": 2300
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 1.0440189775694113,
|
| 673 |
+
"grad_norm": 0.5729214549064636,
|
| 674 |
+
"learning_rate": 0.0009645247864861191,
|
| 675 |
+
"loss": 2.9956,
|
| 676 |
+
"step": 2325
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 1.0552483085820161,
|
| 680 |
+
"grad_norm": 0.5456846356391907,
|
| 681 |
+
"learning_rate": 0.0009636950825074618,
|
| 682 |
+
"loss": 2.985,
|
| 683 |
+
"step": 2350
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 1.0664776395946212,
|
| 687 |
+
"grad_norm": 0.544643223285675,
|
| 688 |
+
"learning_rate": 0.0009628561529956529,
|
| 689 |
+
"loss": 2.9973,
|
| 690 |
+
"step": 2375
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 1.077706970607226,
|
| 694 |
+
"grad_norm": 0.5306060314178467,
|
| 695 |
+
"learning_rate": 0.0009620080146417731,
|
| 696 |
+
"loss": 3.0053,
|
| 697 |
+
"step": 2400
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 1.088936301619831,
|
| 701 |
+
"grad_norm": 0.5001072883605957,
|
| 702 |
+
"learning_rate": 0.0009611506843201193,
|
| 703 |
+
"loss": 3.0244,
|
| 704 |
+
"step": 2425
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 1.1001656326324358,
|
| 708 |
+
"grad_norm": 0.52583909034729,
|
| 709 |
+
"learning_rate": 0.0009602841790878688,
|
| 710 |
+
"loss": 3.0266,
|
| 711 |
+
"step": 2450
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 1.1113949636450409,
|
| 715 |
+
"grad_norm": 0.536445677280426,
|
| 716 |
+
"learning_rate": 0.0009594085161847405,
|
| 717 |
+
"loss": 3.0124,
|
| 718 |
+
"step": 2475
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 1.1226242946576457,
|
| 722 |
+
"grad_norm": 0.5341405272483826,
|
| 723 |
+
"learning_rate": 0.0009585237130326508,
|
| 724 |
+
"loss": 3.0272,
|
| 725 |
+
"step": 2500
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 1.1338536256702507,
|
| 729 |
+
"grad_norm": 0.5340954661369324,
|
| 730 |
+
"learning_rate": 0.0009576297872353686,
|
| 731 |
+
"loss": 3.0152,
|
| 732 |
+
"step": 2525
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 1.1450829566828555,
|
| 736 |
+
"grad_norm": 0.4479193687438965,
|
| 737 |
+
"learning_rate": 0.0009567267565781628,
|
| 738 |
+
"loss": 3.0202,
|
| 739 |
+
"step": 2550
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 1.1563122876954606,
|
| 743 |
+
"grad_norm": 0.5316035747528076,
|
| 744 |
+
"learning_rate": 0.0009558146390274512,
|
| 745 |
+
"loss": 3.015,
|
| 746 |
+
"step": 2575
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 1.1675416187080654,
|
| 750 |
+
"grad_norm": 0.5239371061325073,
|
| 751 |
+
"learning_rate": 0.0009548934527304407,
|
| 752 |
+
"loss": 3.0618,
|
| 753 |
+
"step": 2600
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 1.1787709497206704,
|
| 757 |
+
"grad_norm": 0.6486944556236267,
|
| 758 |
+
"learning_rate": 0.0009539632160147672,
|
| 759 |
+
"loss": 3.0004,
|
| 760 |
+
"step": 2625
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 1.1900002807332752,
|
| 764 |
+
"grad_norm": 0.5308857560157776,
|
| 765 |
+
"learning_rate": 0.0009530239473881313,
|
| 766 |
+
"loss": 3.0425,
|
| 767 |
+
"step": 2650
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 1.2012296117458803,
|
| 771 |
+
"grad_norm": 0.5612149834632874,
|
| 772 |
+
"learning_rate": 0.0009520756655379293,
|
| 773 |
+
"loss": 3.0447,
|
| 774 |
+
"step": 2675
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 1.212458942758485,
|
| 778 |
+
"grad_norm": 0.5429418683052063,
|
| 779 |
+
"learning_rate": 0.0009511183893308821,
|
| 780 |
+
"loss": 2.9887,
|
| 781 |
+
"step": 2700
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 1.2236882737710901,
|
| 785 |
+
"grad_norm": 0.5688816905021667,
|
| 786 |
+
"learning_rate": 0.0009501521378126594,
|
| 787 |
+
"loss": 2.9961,
|
| 788 |
+
"step": 2725
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 1.234917604783695,
|
| 792 |
+
"grad_norm": 0.5409512519836426,
|
| 793 |
+
"learning_rate": 0.0009491769302075008,
|
| 794 |
+
"loss": 3.0,
|
| 795 |
+
"step": 2750
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 1.2461469357963,
|
| 799 |
+
"grad_norm": 0.5384955406188965,
|
| 800 |
+
"learning_rate": 0.0009481927859178337,
|
| 801 |
+
"loss": 3.0271,
|
| 802 |
+
"step": 2775
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 1.2573762668089048,
|
| 806 |
+
"grad_norm": 0.5857961177825928,
|
| 807 |
+
"learning_rate": 0.0009471997245238865,
|
| 808 |
+
"loss": 2.9983,
|
| 809 |
+
"step": 2800
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 1.2686055978215098,
|
| 813 |
+
"grad_norm": 0.5337027907371521,
|
| 814 |
+
"learning_rate": 0.0009461977657833003,
|
| 815 |
+
"loss": 3.0552,
|
| 816 |
+
"step": 2825
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 1.2798349288341146,
|
| 820 |
+
"grad_norm": 0.5078946352005005,
|
| 821 |
+
"learning_rate": 0.0009451869296307341,
|
| 822 |
+
"loss": 3.0191,
|
| 823 |
+
"step": 2850
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 1.2910642598467197,
|
| 827 |
+
"grad_norm": 0.5108660459518433,
|
| 828 |
+
"learning_rate": 0.00094416723617747,
|
| 829 |
+
"loss": 3.0234,
|
| 830 |
+
"step": 2875
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 1.3022935908593245,
|
| 834 |
+
"grad_norm": 0.5631129741668701,
|
| 835 |
+
"learning_rate": 0.0009431387057110118,
|
| 836 |
+
"loss": 3.0319,
|
| 837 |
+
"step": 2900
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 1.3135229218719295,
|
| 841 |
+
"grad_norm": 0.5249589085578918,
|
| 842 |
+
"learning_rate": 0.0009421013586946816,
|
| 843 |
+
"loss": 2.9866,
|
| 844 |
+
"step": 2925
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 1.3247522528845344,
|
| 848 |
+
"grad_norm": 0.4992469251155853,
|
| 849 |
+
"learning_rate": 0.000941055215767213,
|
| 850 |
+
"loss": 3.0144,
|
| 851 |
+
"step": 2950
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 1.3359815838971394,
|
| 855 |
+
"grad_norm": 0.4509263336658478,
|
| 856 |
+
"learning_rate": 0.0009400002977423405,
|
| 857 |
+
"loss": 3.0092,
|
| 858 |
+
"step": 2975
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 1.3472109149097442,
|
| 862 |
+
"grad_norm": 0.515438973903656,
|
| 863 |
+
"learning_rate": 0.0009389366256083849,
|
| 864 |
+
"loss": 2.9993,
|
| 865 |
+
"step": 3000
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 1.3584402459223492,
|
| 869 |
+
"grad_norm": 0.5087840557098389,
|
| 870 |
+
"learning_rate": 0.0009378642205278363,
|
| 871 |
+
"loss": 3.0242,
|
| 872 |
+
"step": 3025
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 1.369669576934954,
|
| 876 |
+
"grad_norm": 0.5046051144599915,
|
| 877 |
+
"learning_rate": 0.0009367831038369326,
|
| 878 |
+
"loss": 2.9971,
|
| 879 |
+
"step": 3050
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 1.380898907947559,
|
| 883 |
+
"grad_norm": 0.5728681087493896,
|
| 884 |
+
"learning_rate": 0.0009356932970452353,
|
| 885 |
+
"loss": 3.0292,
|
| 886 |
+
"step": 3075
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 1.392128238960164,
|
| 890 |
+
"grad_norm": 0.5724380016326904,
|
| 891 |
+
"learning_rate": 0.0009345948218352014,
|
| 892 |
+
"loss": 3.0098,
|
| 893 |
+
"step": 3100
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 1.403357569972769,
|
| 897 |
+
"grad_norm": 0.5322164297103882,
|
| 898 |
+
"learning_rate": 0.0009334877000617518,
|
| 899 |
+
"loss": 2.9968,
|
| 900 |
+
"step": 3125
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 1.4145869009853738,
|
| 904 |
+
"grad_norm": 0.558423638343811,
|
| 905 |
+
"learning_rate": 0.0009323719537518374,
|
| 906 |
+
"loss": 3.0334,
|
| 907 |
+
"step": 3150
|
| 908 |
+
},
|
| 909 |
+
{
|
| 910 |
+
"epoch": 1.4258162319979788,
|
| 911 |
+
"grad_norm": 0.5415078997612,
|
| 912 |
+
"learning_rate": 0.0009312476051039994,
|
| 913 |
+
"loss": 3.0313,
|
| 914 |
+
"step": 3175
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 1.4370455630105836,
|
| 918 |
+
"grad_norm": 0.46919873356819153,
|
| 919 |
+
"learning_rate": 0.0009301146764879292,
|
| 920 |
+
"loss": 2.9992,
|
| 921 |
+
"step": 3200
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 1.4482748940231884,
|
| 925 |
+
"grad_norm": 0.5965465903282166,
|
| 926 |
+
"learning_rate": 0.0009289731904440217,
|
| 927 |
+
"loss": 3.0071,
|
| 928 |
+
"step": 3225
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 1.4595042250357935,
|
| 932 |
+
"grad_norm": 0.4882059693336487,
|
| 933 |
+
"learning_rate": 0.0009278231696829288,
|
| 934 |
+
"loss": 2.968,
|
| 935 |
+
"step": 3250
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 1.4707335560483985,
|
| 939 |
+
"grad_norm": 0.6297493577003479,
|
| 940 |
+
"learning_rate": 0.0009266646370851055,
|
| 941 |
+
"loss": 3.0411,
|
| 942 |
+
"step": 3275
|
| 943 |
+
},
|
| 944 |
+
{
|
| 945 |
+
"epoch": 1.4819628870610033,
|
| 946 |
+
"grad_norm": 0.5603842735290527,
|
| 947 |
+
"learning_rate": 0.0009254976157003563,
|
| 948 |
+
"loss": 3.0203,
|
| 949 |
+
"step": 3300
|
| 950 |
+
},
|
| 951 |
+
{
|
| 952 |
+
"epoch": 1.4931922180736081,
|
| 953 |
+
"grad_norm": 0.49509698152542114,
|
| 954 |
+
"learning_rate": 0.0009243221287473755,
|
| 955 |
+
"loss": 3.0176,
|
| 956 |
+
"step": 3325
|
| 957 |
+
},
|
| 958 |
+
{
|
| 959 |
+
"epoch": 1.5044215490862132,
|
| 960 |
+
"grad_norm": 0.48536983132362366,
|
| 961 |
+
"learning_rate": 0.0009231381996132862,
|
| 962 |
+
"loss": 2.9547,
|
| 963 |
+
"step": 3350
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 1.5156508800988182,
|
| 967 |
+
"grad_norm": 0.47351208329200745,
|
| 968 |
+
"learning_rate": 0.0009219458518531739,
|
| 969 |
+
"loss": 2.9666,
|
| 970 |
+
"step": 3375
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 1.526880211111423,
|
| 974 |
+
"grad_norm": 0.5615521669387817,
|
| 975 |
+
"learning_rate": 0.0009207451091896191,
|
| 976 |
+
"loss": 3.0295,
|
| 977 |
+
"step": 3400
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 1.5381095421240278,
|
| 981 |
+
"grad_norm": 0.5138916969299316,
|
| 982 |
+
"learning_rate": 0.0009195359955122244,
|
| 983 |
+
"loss": 3.0146,
|
| 984 |
+
"step": 3425
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 1.5493388731366329,
|
| 988 |
+
"grad_norm": 0.5883649587631226,
|
| 989 |
+
"learning_rate": 0.0009183185348771392,
|
| 990 |
+
"loss": 3.0151,
|
| 991 |
+
"step": 3450
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 1.560568204149238,
|
| 995 |
+
"grad_norm": 0.5921751260757446,
|
| 996 |
+
"learning_rate": 0.0009170927515065821,
|
| 997 |
+
"loss": 3.0314,
|
| 998 |
+
"step": 3475
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 1.5717975351618427,
|
| 1002 |
+
"grad_norm": 0.5592530965805054,
|
| 1003 |
+
"learning_rate": 0.0009158586697883576,
|
| 1004 |
+
"loss": 2.9921,
|
| 1005 |
+
"step": 3500
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 1.5830268661744475,
|
| 1009 |
+
"grad_norm": 0.5621814727783203,
|
| 1010 |
+
"learning_rate": 0.0009146163142753716,
|
| 1011 |
+
"loss": 2.9987,
|
| 1012 |
+
"step": 3525
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 1.5942561971870526,
|
| 1016 |
+
"grad_norm": 0.5482603311538696,
|
| 1017 |
+
"learning_rate": 0.0009133657096851431,
|
| 1018 |
+
"loss": 2.9802,
|
| 1019 |
+
"step": 3550
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 1.6054855281996576,
|
| 1023 |
+
"grad_norm": 0.5254377722740173,
|
| 1024 |
+
"learning_rate": 0.0009121068808993124,
|
| 1025 |
+
"loss": 3.0121,
|
| 1026 |
+
"step": 3575
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 1.6167148592122624,
|
| 1030 |
+
"grad_norm": 0.47623664140701294,
|
| 1031 |
+
"learning_rate": 0.0009108398529631451,
|
| 1032 |
+
"loss": 3.0068,
|
| 1033 |
+
"step": 3600
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 1.6279441902248672,
|
| 1037 |
+
"grad_norm": 0.49733710289001465,
|
| 1038 |
+
"learning_rate": 0.0009095646510850351,
|
| 1039 |
+
"loss": 3.0104,
|
| 1040 |
+
"step": 3625
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 1.6391735212374723,
|
| 1044 |
+
"grad_norm": 0.5388875603675842,
|
| 1045 |
+
"learning_rate": 0.0009082813006360026,
|
| 1046 |
+
"loss": 2.9823,
|
| 1047 |
+
"step": 3650
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"epoch": 1.6504028522500773,
|
| 1051 |
+
"grad_norm": 0.5329872965812683,
|
| 1052 |
+
"learning_rate": 0.0009069898271491887,
|
| 1053 |
+
"loss": 2.9945,
|
| 1054 |
+
"step": 3675
|
| 1055 |
+
},
|
| 1056 |
+
{
|
| 1057 |
+
"epoch": 1.6616321832626821,
|
| 1058 |
+
"grad_norm": 0.5175071358680725,
|
| 1059 |
+
"learning_rate": 0.0009056902563193486,
|
| 1060 |
+
"loss": 2.9875,
|
| 1061 |
+
"step": 3700
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 1.672861514275287,
|
| 1065 |
+
"grad_norm": 0.514216423034668,
|
| 1066 |
+
"learning_rate": 0.0009043826140023388,
|
| 1067 |
+
"loss": 3.016,
|
| 1068 |
+
"step": 3725
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 1.684090845287892,
|
| 1072 |
+
"grad_norm": 0.5547803640365601,
|
| 1073 |
+
"learning_rate": 0.0009030669262146046,
|
| 1074 |
+
"loss": 2.9906,
|
| 1075 |
+
"step": 3750
|
| 1076 |
+
},
|
| 1077 |
+
{
|
| 1078 |
+
"epoch": 1.695320176300497,
|
| 1079 |
+
"grad_norm": 0.5035697817802429,
|
| 1080 |
+
"learning_rate": 0.0009017432191326611,
|
| 1081 |
+
"loss": 2.9795,
|
| 1082 |
+
"step": 3775
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 1.7065495073131018,
|
| 1086 |
+
"grad_norm": 0.4960135519504547,
|
| 1087 |
+
"learning_rate": 0.0009004115190925724,
|
| 1088 |
+
"loss": 2.986,
|
| 1089 |
+
"step": 3800
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 1.7177788383257067,
|
| 1093 |
+
"grad_norm": 0.5573786497116089,
|
| 1094 |
+
"learning_rate": 0.0008990718525894286,
|
| 1095 |
+
"loss": 2.9981,
|
| 1096 |
+
"step": 3825
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 1.7290081693383117,
|
| 1100 |
+
"grad_norm": 0.558542788028717,
|
| 1101 |
+
"learning_rate": 0.0008977242462768177,
|
| 1102 |
+
"loss": 3.0122,
|
| 1103 |
+
"step": 3850
|
| 1104 |
+
},
|
| 1105 |
+
{
|
| 1106 |
+
"epoch": 1.7402375003509167,
|
| 1107 |
+
"grad_norm": 0.48205050826072693,
|
| 1108 |
+
"learning_rate": 0.0008963687269662957,
|
| 1109 |
+
"loss": 2.9558,
|
| 1110 |
+
"step": 3875
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"epoch": 1.7514668313635215,
|
| 1114 |
+
"grad_norm": 0.48525846004486084,
|
| 1115 |
+
"learning_rate": 0.0008950053216268534,
|
| 1116 |
+
"loss": 3.0034,
|
| 1117 |
+
"step": 3900
|
| 1118 |
+
},
|
| 1119 |
+
{
|
| 1120 |
+
"epoch": 1.7626961623761264,
|
| 1121 |
+
"grad_norm": 0.5863490700721741,
|
| 1122 |
+
"learning_rate": 0.0008936340573843795,
|
| 1123 |
+
"loss": 3.0222,
|
| 1124 |
+
"step": 3925
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 1.7739254933887314,
|
| 1128 |
+
"grad_norm": 0.54740309715271,
|
| 1129 |
+
"learning_rate": 0.0008922549615211206,
|
| 1130 |
+
"loss": 2.9785,
|
| 1131 |
+
"step": 3950
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 1.7851548244013364,
|
| 1135 |
+
"grad_norm": 0.5275555849075317,
|
| 1136 |
+
"learning_rate": 0.0008908680614751392,
|
| 1137 |
+
"loss": 2.982,
|
| 1138 |
+
"step": 3975
|
| 1139 |
+
},
|
| 1140 |
+
{
|
| 1141 |
+
"epoch": 1.7963841554139413,
|
| 1142 |
+
"grad_norm": 0.5472078919410706,
|
| 1143 |
+
"learning_rate": 0.0008894733848397674,
|
| 1144 |
+
"loss": 3.0128,
|
| 1145 |
+
"step": 4000
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 1.807613486426546,
|
| 1149 |
+
"grad_norm": 0.5604407787322998,
|
| 1150 |
+
"learning_rate": 0.0008880709593630578,
|
| 1151 |
+
"loss": 3.0119,
|
| 1152 |
+
"step": 4025
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 1.818842817439151,
|
| 1156 |
+
"grad_norm": 0.5137823224067688,
|
| 1157 |
+
"learning_rate": 0.0008866608129472313,
|
| 1158 |
+
"loss": 2.9858,
|
| 1159 |
+
"step": 4050
|
| 1160 |
+
},
|
| 1161 |
+
{
|
| 1162 |
+
"epoch": 1.8300721484517561,
|
| 1163 |
+
"grad_norm": 0.5707024931907654,
|
| 1164 |
+
"learning_rate": 0.0008852429736481227,
|
| 1165 |
+
"loss": 3.013,
|
| 1166 |
+
"step": 4075
|
| 1167 |
+
},
|
| 1168 |
+
{
|
| 1169 |
+
"epoch": 1.841301479464361,
|
| 1170 |
+
"grad_norm": 0.5344915986061096,
|
| 1171 |
+
"learning_rate": 0.0008838174696746215,
|
| 1172 |
+
"loss": 2.9899,
|
| 1173 |
+
"step": 4100
|
| 1174 |
+
},
|
| 1175 |
+
{
|
| 1176 |
+
"epoch": 1.8525308104769658,
|
| 1177 |
+
"grad_norm": 0.504295289516449,
|
| 1178 |
+
"learning_rate": 0.0008823843293881117,
|
| 1179 |
+
"loss": 3.0095,
|
| 1180 |
+
"step": 4125
|
| 1181 |
+
},
|
| 1182 |
+
{
|
| 1183 |
+
"epoch": 1.8637601414895708,
|
| 1184 |
+
"grad_norm": 0.5654752254486084,
|
| 1185 |
+
"learning_rate": 0.0008809435813019065,
|
| 1186 |
+
"loss": 2.988,
|
| 1187 |
+
"step": 4150
|
| 1188 |
+
},
|
| 1189 |
+
{
|
| 1190 |
+
"epoch": 1.8749894725021758,
|
| 1191 |
+
"grad_norm": 0.5086371302604675,
|
| 1192 |
+
"learning_rate": 0.0008794952540806817,
|
| 1193 |
+
"loss": 3.0304,
|
| 1194 |
+
"step": 4175
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 1.8862188035147807,
|
| 1198 |
+
"grad_norm": 0.5218560099601746,
|
| 1199 |
+
"learning_rate": 0.0008780393765399055,
|
| 1200 |
+
"loss": 2.9817,
|
| 1201 |
+
"step": 4200
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"epoch": 1.8974481345273855,
|
| 1205 |
+
"grad_norm": 0.48831528425216675,
|
| 1206 |
+
"learning_rate": 0.0008765759776452646,
|
| 1207 |
+
"loss": 3.0245,
|
| 1208 |
+
"step": 4225
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 1.9086774655399905,
|
| 1212 |
+
"grad_norm": 0.5015767812728882,
|
| 1213 |
+
"learning_rate": 0.0008751050865120882,
|
| 1214 |
+
"loss": 3.0238,
|
| 1215 |
+
"step": 4250
|
| 1216 |
+
},
|
| 1217 |
+
{
|
| 1218 |
+
"epoch": 1.9199067965525953,
|
| 1219 |
+
"grad_norm": 0.5325757265090942,
|
| 1220 |
+
"learning_rate": 0.000873626732404769,
|
| 1221 |
+
"loss": 2.9993,
|
| 1222 |
+
"step": 4275
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"epoch": 1.9311361275652001,
|
| 1226 |
+
"grad_norm": 0.48629334568977356,
|
| 1227 |
+
"learning_rate": 0.0008721409447361803,
|
| 1228 |
+
"loss": 2.9634,
|
| 1229 |
+
"step": 4300
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 1.9423654585778052,
|
| 1233 |
+
"grad_norm": 0.49022358655929565,
|
| 1234 |
+
"learning_rate": 0.0008706477530670917,
|
| 1235 |
+
"loss": 2.9736,
|
| 1236 |
+
"step": 4325
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 1.9535947895904102,
|
| 1240 |
+
"grad_norm": 0.5039647221565247,
|
| 1241 |
+
"learning_rate": 0.0008691471871055801,
|
| 1242 |
+
"loss": 2.9802,
|
| 1243 |
+
"step": 4350
|
| 1244 |
+
},
|
| 1245 |
+
{
|
| 1246 |
+
"epoch": 1.964824120603015,
|
| 1247 |
+
"grad_norm": 0.5223824977874756,
|
| 1248 |
+
"learning_rate": 0.0008676392767064391,
|
| 1249 |
+
"loss": 3.0397,
|
| 1250 |
+
"step": 4375
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 1.9760534516156198,
|
| 1254 |
+
"grad_norm": 0.5172558426856995,
|
| 1255 |
+
"learning_rate": 0.0008661240518705854,
|
| 1256 |
+
"loss": 2.9756,
|
| 1257 |
+
"step": 4400
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 1.9872827826282249,
|
| 1261 |
+
"grad_norm": 0.4955403208732605,
|
| 1262 |
+
"learning_rate": 0.0008646015427444609,
|
| 1263 |
+
"loss": 2.9748,
|
| 1264 |
+
"step": 4425
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 1.99851211364083,
|
| 1268 |
+
"grad_norm": 0.553287923336029,
|
| 1269 |
+
"learning_rate": 0.0008630717796194337,
|
| 1270 |
+
"loss": 2.9501,
|
| 1271 |
+
"step": 4450
|
| 1272 |
+
},
|
| 1273 |
+
{
|
| 1274 |
+
"epoch": 2.0,
|
| 1275 |
+
"eval_loss": 3.1917288303375244,
|
| 1276 |
+
"eval_runtime": 227.8959,
|
| 1277 |
+
"eval_samples_per_second": 55.218,
|
| 1278 |
+
"eval_steps_per_second": 55.218,
|
| 1279 |
+
"step": 4454
|
| 1280 |
+
},
|
| 1281 |
+
{
|
| 1282 |
+
"epoch": 2.0094326380505882,
|
| 1283 |
+
"grad_norm": 0.5545716285705566,
|
| 1284 |
+
"learning_rate": 0.0008615347929311949,
|
| 1285 |
+
"loss": 2.7426,
|
| 1286 |
+
"step": 4475
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 2.020661969063193,
|
| 1290 |
+
"grad_norm": 0.5116350650787354,
|
| 1291 |
+
"learning_rate": 0.0008599906132591541,
|
| 1292 |
+
"loss": 2.6669,
|
| 1293 |
+
"step": 4500
|
| 1294 |
+
},
|
| 1295 |
+
{
|
| 1296 |
+
"epoch": 2.031891300075798,
|
| 1297 |
+
"grad_norm": 0.6248686909675598,
|
| 1298 |
+
"learning_rate": 0.0008584392713258295,
|
| 1299 |
+
"loss": 2.6597,
|
| 1300 |
+
"step": 4525
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"epoch": 2.043120631088403,
|
| 1304 |
+
"grad_norm": 0.5305931568145752,
|
| 1305 |
+
"learning_rate": 0.0008568807979962379,
|
| 1306 |
+
"loss": 2.6635,
|
| 1307 |
+
"step": 4550
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 2.054349962101008,
|
| 1311 |
+
"grad_norm": 0.5366395711898804,
|
| 1312 |
+
"learning_rate": 0.0008553152242772798,
|
| 1313 |
+
"loss": 2.668,
|
| 1314 |
+
"step": 4575
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 2.0655792931136125,
|
| 1318 |
+
"grad_norm": 0.6074578762054443,
|
| 1319 |
+
"learning_rate": 0.0008537425813171232,
|
| 1320 |
+
"loss": 2.7031,
|
| 1321 |
+
"step": 4600
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 2.0768086241262176,
|
| 1325 |
+
"grad_norm": 0.6074210405349731,
|
| 1326 |
+
"learning_rate": 0.0008521629004045832,
|
| 1327 |
+
"loss": 2.6721,
|
| 1328 |
+
"step": 4625
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 2.0880379551388226,
|
| 1332 |
+
"grad_norm": 0.6065968871116638,
|
| 1333 |
+
"learning_rate": 0.0008505762129685002,
|
| 1334 |
+
"loss": 2.6774,
|
| 1335 |
+
"step": 4650
|
| 1336 |
+
},
|
| 1337 |
+
{
|
| 1338 |
+
"epoch": 2.0992672861514277,
|
| 1339 |
+
"grad_norm": 0.5635676383972168,
|
| 1340 |
+
"learning_rate": 0.0008489825505771136,
|
| 1341 |
+
"loss": 2.6537,
|
| 1342 |
+
"step": 4675
|
| 1343 |
+
},
|
| 1344 |
+
{
|
| 1345 |
+
"epoch": 2.1104966171640323,
|
| 1346 |
+
"grad_norm": 0.5447210669517517,
|
| 1347 |
+
"learning_rate": 0.000847381944937435,
|
| 1348 |
+
"loss": 2.6964,
|
| 1349 |
+
"step": 4700
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"epoch": 2.1217259481766373,
|
| 1353 |
+
"grad_norm": 0.6450474858283997,
|
| 1354 |
+
"learning_rate": 0.0008457744278946162,
|
| 1355 |
+
"loss": 2.6591,
|
| 1356 |
+
"step": 4725
|
| 1357 |
+
},
|
| 1358 |
+
{
|
| 1359 |
+
"epoch": 2.1329552791892423,
|
| 1360 |
+
"grad_norm": 0.5620629787445068,
|
| 1361 |
+
"learning_rate": 0.0008441600314313165,
|
| 1362 |
+
"loss": 2.6787,
|
| 1363 |
+
"step": 4750
|
| 1364 |
+
},
|
| 1365 |
+
{
|
| 1366 |
+
"epoch": 2.1441846102018474,
|
| 1367 |
+
"grad_norm": 0.5661433935165405,
|
| 1368 |
+
"learning_rate": 0.0008425387876670658,
|
| 1369 |
+
"loss": 2.7193,
|
| 1370 |
+
"step": 4775
|
| 1371 |
+
},
|
| 1372 |
+
{
|
| 1373 |
+
"epoch": 2.155413941214452,
|
| 1374 |
+
"grad_norm": 0.6069077849388123,
|
| 1375 |
+
"learning_rate": 0.0008409107288576259,
|
| 1376 |
+
"loss": 2.6947,
|
| 1377 |
+
"step": 4800
|
| 1378 |
+
},
|
| 1379 |
+
{
|
| 1380 |
+
"epoch": 2.166643272227057,
|
| 1381 |
+
"grad_norm": 0.6271758675575256,
|
| 1382 |
+
"learning_rate": 0.0008392758873943484,
|
| 1383 |
+
"loss": 2.6952,
|
| 1384 |
+
"step": 4825
|
| 1385 |
+
},
|
| 1386 |
+
{
|
| 1387 |
+
"epoch": 2.177872603239662,
|
| 1388 |
+
"grad_norm": 0.5473480820655823,
|
| 1389 |
+
"learning_rate": 0.0008376342958035308,
|
| 1390 |
+
"loss": 2.6981,
|
| 1391 |
+
"step": 4850
|
| 1392 |
+
},
|
| 1393 |
+
{
|
| 1394 |
+
"epoch": 2.189101934252267,
|
| 1395 |
+
"grad_norm": 0.631519079208374,
|
| 1396 |
+
"learning_rate": 0.0008359859867457686,
|
| 1397 |
+
"loss": 2.6921,
|
| 1398 |
+
"step": 4875
|
| 1399 |
+
},
|
| 1400 |
+
{
|
| 1401 |
+
"epoch": 2.2003312652648717,
|
| 1402 |
+
"grad_norm": 0.6639063954353333,
|
| 1403 |
+
"learning_rate": 0.0008343309930153064,
|
| 1404 |
+
"loss": 2.6837,
|
| 1405 |
+
"step": 4900
|
| 1406 |
+
},
|
| 1407 |
+
{
|
| 1408 |
+
"epoch": 2.2115605962774767,
|
| 1409 |
+
"grad_norm": 0.6233927011489868,
|
| 1410 |
+
"learning_rate": 0.0008326693475393846,
|
| 1411 |
+
"loss": 2.7112,
|
| 1412 |
+
"step": 4925
|
| 1413 |
+
},
|
| 1414 |
+
{
|
| 1415 |
+
"epoch": 2.2227899272900817,
|
| 1416 |
+
"grad_norm": 0.636464536190033,
|
| 1417 |
+
"learning_rate": 0.0008310010833775849,
|
| 1418 |
+
"loss": 2.7213,
|
| 1419 |
+
"step": 4950
|
| 1420 |
+
},
|
| 1421 |
+
{
|
| 1422 |
+
"epoch": 2.2340192583026868,
|
| 1423 |
+
"grad_norm": 0.5854448676109314,
|
| 1424 |
+
"learning_rate": 0.0008293262337211723,
|
| 1425 |
+
"loss": 2.7131,
|
| 1426 |
+
"step": 4975
|
| 1427 |
+
},
|
| 1428 |
+
{
|
| 1429 |
+
"epoch": 2.2452485893152914,
|
| 1430 |
+
"grad_norm": 0.6958891749382019,
|
| 1431 |
+
"learning_rate": 0.0008276448318924346,
|
| 1432 |
+
"loss": 2.6883,
|
| 1433 |
+
"step": 5000
|
| 1434 |
+
},
|
| 1435 |
+
{
|
| 1436 |
+
"epoch": 2.2564779203278964,
|
| 1437 |
+
"grad_norm": 0.5899659991264343,
|
| 1438 |
+
"learning_rate": 0.0008259569113440198,
|
| 1439 |
+
"loss": 2.6872,
|
| 1440 |
+
"step": 5025
|
| 1441 |
+
},
|
| 1442 |
+
{
|
| 1443 |
+
"epoch": 2.2677072513405014,
|
| 1444 |
+
"grad_norm": 0.6791245937347412,
|
| 1445 |
+
"learning_rate": 0.0008242625056582698,
|
| 1446 |
+
"loss": 2.7202,
|
| 1447 |
+
"step": 5050
|
| 1448 |
+
},
|
| 1449 |
+
{
|
| 1450 |
+
"epoch": 2.2789365823531065,
|
| 1451 |
+
"grad_norm": 0.5778390169143677,
|
| 1452 |
+
"learning_rate": 0.0008225616485465535,
|
| 1453 |
+
"loss": 2.7153,
|
| 1454 |
+
"step": 5075
|
| 1455 |
+
},
|
| 1456 |
+
{
|
| 1457 |
+
"epoch": 2.290165913365711,
|
| 1458 |
+
"grad_norm": 0.5727918148040771,
|
| 1459 |
+
"learning_rate": 0.000820854373848595,
|
| 1460 |
+
"loss": 2.7314,
|
| 1461 |
+
"step": 5100
|
| 1462 |
+
},
|
| 1463 |
+
{
|
| 1464 |
+
"epoch": 2.301395244378316,
|
| 1465 |
+
"grad_norm": 0.6461373567581177,
|
| 1466 |
+
"learning_rate": 0.0008191407155318007,
|
| 1467 |
+
"loss": 2.6973,
|
| 1468 |
+
"step": 5125
|
| 1469 |
+
},
|
| 1470 |
+
{
|
| 1471 |
+
"epoch": 2.312624575390921,
|
| 1472 |
+
"grad_norm": 0.6795935034751892,
|
| 1473 |
+
"learning_rate": 0.0008174207076905835,
|
| 1474 |
+
"loss": 2.6605,
|
| 1475 |
+
"step": 5150
|
| 1476 |
+
},
|
| 1477 |
+
{
|
| 1478 |
+
"epoch": 2.323853906403526,
|
| 1479 |
+
"grad_norm": 0.6217265725135803,
|
| 1480 |
+
"learning_rate": 0.0008156943845456843,
|
| 1481 |
+
"loss": 2.6715,
|
| 1482 |
+
"step": 5175
|
| 1483 |
+
},
|
| 1484 |
+
{
|
| 1485 |
+
"epoch": 2.3350832374161308,
|
| 1486 |
+
"grad_norm": 0.6071234941482544,
|
| 1487 |
+
"learning_rate": 0.0008139617804434918,
|
| 1488 |
+
"loss": 2.6806,
|
| 1489 |
+
"step": 5200
|
| 1490 |
+
},
|
| 1491 |
+
{
|
| 1492 |
+
"epoch": 2.346312568428736,
|
| 1493 |
+
"grad_norm": 0.6218218207359314,
|
| 1494 |
+
"learning_rate": 0.0008122229298553583,
|
| 1495 |
+
"loss": 2.7077,
|
| 1496 |
+
"step": 5225
|
| 1497 |
+
},
|
| 1498 |
+
{
|
| 1499 |
+
"epoch": 2.357541899441341,
|
| 1500 |
+
"grad_norm": 0.5912306904792786,
|
| 1501 |
+
"learning_rate": 0.0008104778673769142,
|
| 1502 |
+
"loss": 2.7314,
|
| 1503 |
+
"step": 5250
|
| 1504 |
+
},
|
| 1505 |
+
{
|
| 1506 |
+
"epoch": 2.368771230453946,
|
| 1507 |
+
"grad_norm": 0.5841456651687622,
|
| 1508 |
+
"learning_rate": 0.0008087266277273799,
|
| 1509 |
+
"loss": 2.6645,
|
| 1510 |
+
"step": 5275
|
| 1511 |
+
},
|
| 1512 |
+
{
|
| 1513 |
+
"epoch": 2.3800005614665505,
|
| 1514 |
+
"grad_norm": 0.6491414904594421,
|
| 1515 |
+
"learning_rate": 0.0008069692457488749,
|
| 1516 |
+
"loss": 2.7115,
|
| 1517 |
+
"step": 5300
|
| 1518 |
+
},
|
| 1519 |
+
{
|
| 1520 |
+
"epoch": 2.3912298924791555,
|
| 1521 |
+
"grad_norm": 0.6534895896911621,
|
| 1522 |
+
"learning_rate": 0.0008052057564057244,
|
| 1523 |
+
"loss": 2.7057,
|
| 1524 |
+
"step": 5325
|
| 1525 |
+
},
|
| 1526 |
+
{
|
| 1527 |
+
"epoch": 2.4024592234917606,
|
| 1528 |
+
"grad_norm": 0.5900655388832092,
|
| 1529 |
+
"learning_rate": 0.000803436194783764,
|
| 1530 |
+
"loss": 2.7302,
|
| 1531 |
+
"step": 5350
|
| 1532 |
+
},
|
| 1533 |
+
{
|
| 1534 |
+
"epoch": 2.4136885545043656,
|
| 1535 |
+
"grad_norm": 0.5586231350898743,
|
| 1536 |
+
"learning_rate": 0.0008016605960896412,
|
| 1537 |
+
"loss": 2.7339,
|
| 1538 |
+
"step": 5375
|
| 1539 |
+
},
|
| 1540 |
+
{
|
| 1541 |
+
"epoch": 2.42491788551697,
|
| 1542 |
+
"grad_norm": 0.705515444278717,
|
| 1543 |
+
"learning_rate": 0.0007998789956501159,
|
| 1544 |
+
"loss": 2.7323,
|
| 1545 |
+
"step": 5400
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 2.436147216529575,
|
| 1549 |
+
"grad_norm": 0.5936200022697449,
|
| 1550 |
+
"learning_rate": 0.0007980914289113558,
|
| 1551 |
+
"loss": 2.7116,
|
| 1552 |
+
"step": 5425
|
| 1553 |
+
},
|
| 1554 |
+
{
|
| 1555 |
+
"epoch": 2.4473765475421803,
|
| 1556 |
+
"grad_norm": 0.6085701584815979,
|
| 1557 |
+
"learning_rate": 0.000796297931438233,
|
| 1558 |
+
"loss": 2.7406,
|
| 1559 |
+
"step": 5450
|
| 1560 |
+
},
|
| 1561 |
+
{
|
| 1562 |
+
"epoch": 2.458605878554785,
|
| 1563 |
+
"grad_norm": 0.5549573302268982,
|
| 1564 |
+
"learning_rate": 0.0007944985389136157,
|
| 1565 |
+
"loss": 2.7408,
|
| 1566 |
+
"step": 5475
|
| 1567 |
+
},
|
| 1568 |
+
{
|
| 1569 |
+
"epoch": 2.46983520956739,
|
| 1570 |
+
"grad_norm": 0.5694999694824219,
|
| 1571 |
+
"learning_rate": 0.0007926932871376575,
|
| 1572 |
+
"loss": 2.7216,
|
| 1573 |
+
"step": 5500
|
| 1574 |
+
},
|
| 1575 |
+
{
|
| 1576 |
+
"epoch": 2.481064540579995,
|
| 1577 |
+
"grad_norm": 0.5795106887817383,
|
| 1578 |
+
"learning_rate": 0.0007908822120270867,
|
| 1579 |
+
"loss": 2.6724,
|
| 1580 |
+
"step": 5525
|
| 1581 |
+
},
|
| 1582 |
+
{
|
| 1583 |
+
"epoch": 2.4922938715926,
|
| 1584 |
+
"grad_norm": 0.5619019865989685,
|
| 1585 |
+
"learning_rate": 0.0007890653496144902,
|
| 1586 |
+
"loss": 2.6867,
|
| 1587 |
+
"step": 5550
|
| 1588 |
+
},
|
| 1589 |
+
{
|
| 1590 |
+
"epoch": 2.503523202605205,
|
| 1591 |
+
"grad_norm": 0.5836601257324219,
|
| 1592 |
+
"learning_rate": 0.0007872427360475974,
|
| 1593 |
+
"loss": 2.7091,
|
| 1594 |
+
"step": 5575
|
| 1595 |
+
},
|
| 1596 |
+
{
|
| 1597 |
+
"epoch": 2.5147525336178096,
|
| 1598 |
+
"grad_norm": 0.6521953344345093,
|
| 1599 |
+
"learning_rate": 0.0007854144075885614,
|
| 1600 |
+
"loss": 2.7138,
|
| 1601 |
+
"step": 5600
|
| 1602 |
+
},
|
| 1603 |
+
{
|
| 1604 |
+
"epoch": 2.5259818646304146,
|
| 1605 |
+
"grad_norm": 0.6129563450813293,
|
| 1606 |
+
"learning_rate": 0.0007835804006132364,
|
| 1607 |
+
"loss": 2.6796,
|
| 1608 |
+
"step": 5625
|
| 1609 |
+
},
|
| 1610 |
+
{
|
| 1611 |
+
"epoch": 2.5372111956430197,
|
| 1612 |
+
"grad_norm": 0.5933245420455933,
|
| 1613 |
+
"learning_rate": 0.0007817407516104547,
|
| 1614 |
+
"loss": 2.6541,
|
| 1615 |
+
"step": 5650
|
| 1616 |
+
},
|
| 1617 |
+
{
|
| 1618 |
+
"epoch": 2.5484405266556243,
|
| 1619 |
+
"grad_norm": 0.5935245156288147,
|
| 1620 |
+
"learning_rate": 0.0007798954971813009,
|
| 1621 |
+
"loss": 2.6849,
|
| 1622 |
+
"step": 5675
|
| 1623 |
+
},
|
| 1624 |
+
{
|
| 1625 |
+
"epoch": 2.5596698576682293,
|
| 1626 |
+
"grad_norm": 0.7134594321250916,
|
| 1627 |
+
"learning_rate": 0.0007780446740383829,
|
| 1628 |
+
"loss": 2.7141,
|
| 1629 |
+
"step": 5700
|
| 1630 |
+
},
|
| 1631 |
+
{
|
| 1632 |
+
"epoch": 2.5708991886808343,
|
| 1633 |
+
"grad_norm": 0.6013050675392151,
|
| 1634 |
+
"learning_rate": 0.0007761883190051029,
|
| 1635 |
+
"loss": 2.7276,
|
| 1636 |
+
"step": 5725
|
| 1637 |
+
},
|
| 1638 |
+
{
|
| 1639 |
+
"epoch": 2.5821285196934394,
|
| 1640 |
+
"grad_norm": 0.6081655025482178,
|
| 1641 |
+
"learning_rate": 0.000774326469014923,
|
| 1642 |
+
"loss": 2.7205,
|
| 1643 |
+
"step": 5750
|
| 1644 |
+
},
|
| 1645 |
+
{
|
| 1646 |
+
"epoch": 2.5933578507060444,
|
| 1647 |
+
"grad_norm": 0.6464730501174927,
|
| 1648 |
+
"learning_rate": 0.0007724591611106315,
|
| 1649 |
+
"loss": 2.6872,
|
| 1650 |
+
"step": 5775
|
| 1651 |
+
},
|
| 1652 |
+
{
|
| 1653 |
+
"epoch": 2.604587181718649,
|
| 1654 |
+
"grad_norm": 0.578700840473175,
|
| 1655 |
+
"learning_rate": 0.0007705864324436059,
|
| 1656 |
+
"loss": 2.7152,
|
| 1657 |
+
"step": 5800
|
| 1658 |
+
},
|
| 1659 |
+
{
|
| 1660 |
+
"epoch": 2.615816512731254,
|
| 1661 |
+
"grad_norm": 0.5782270431518555,
|
| 1662 |
+
"learning_rate": 0.000768708320273073,
|
| 1663 |
+
"loss": 2.7233,
|
| 1664 |
+
"step": 5825
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"epoch": 2.627045843743859,
|
| 1668 |
+
"grad_norm": 0.5789017081260681,
|
| 1669 |
+
"learning_rate": 0.000766824861965369,
|
| 1670 |
+
"loss": 2.7474,
|
| 1671 |
+
"step": 5850
|
| 1672 |
+
},
|
| 1673 |
+
{
|
| 1674 |
+
"epoch": 2.6382751747564637,
|
| 1675 |
+
"grad_norm": 0.6152642369270325,
|
| 1676 |
+
"learning_rate": 0.0007649360949931941,
|
| 1677 |
+
"loss": 2.7071,
|
| 1678 |
+
"step": 5875
|
| 1679 |
+
},
|
| 1680 |
+
{
|
| 1681 |
+
"epoch": 2.6495045057690687,
|
| 1682 |
+
"grad_norm": 0.6417413353919983,
|
| 1683 |
+
"learning_rate": 0.0007630420569348688,
|
| 1684 |
+
"loss": 2.694,
|
| 1685 |
+
"step": 5900
|
| 1686 |
+
},
|
| 1687 |
+
{
|
| 1688 |
+
"epoch": 2.6607338367816737,
|
| 1689 |
+
"grad_norm": 0.5956742167472839,
|
| 1690 |
+
"learning_rate": 0.0007611427854735855,
|
| 1691 |
+
"loss": 2.7318,
|
| 1692 |
+
"step": 5925
|
| 1693 |
+
},
|
| 1694 |
+
{
|
| 1695 |
+
"epoch": 2.671963167794279,
|
| 1696 |
+
"grad_norm": 0.6362649202346802,
|
| 1697 |
+
"learning_rate": 0.0007592383183966581,
|
| 1698 |
+
"loss": 2.6966,
|
| 1699 |
+
"step": 5950
|
| 1700 |
+
},
|
| 1701 |
+
{
|
| 1702 |
+
"epoch": 2.683192498806884,
|
| 1703 |
+
"grad_norm": 0.5551230311393738,
|
| 1704 |
+
"learning_rate": 0.0007573286935947715,
|
| 1705 |
+
"loss": 2.6876,
|
| 1706 |
+
"step": 5975
|
| 1707 |
+
},
|
| 1708 |
+
{
|
| 1709 |
+
"epoch": 2.6944218298194884,
|
| 1710 |
+
"grad_norm": 0.6288260817527771,
|
| 1711 |
+
"learning_rate": 0.0007554139490612269,
|
| 1712 |
+
"loss": 2.7336,
|
| 1713 |
+
"step": 6000
|
| 1714 |
+
},
|
| 1715 |
+
{
|
| 1716 |
+
"epoch": 2.7056511608320934,
|
| 1717 |
+
"grad_norm": 0.5820605158805847,
|
| 1718 |
+
"learning_rate": 0.0007534941228911856,
|
| 1719 |
+
"loss": 2.683,
|
| 1720 |
+
"step": 6025
|
| 1721 |
+
},
|
| 1722 |
+
{
|
| 1723 |
+
"epoch": 2.7168804918446985,
|
| 1724 |
+
"grad_norm": 0.6264435648918152,
|
| 1725 |
+
"learning_rate": 0.0007515692532809126,
|
| 1726 |
+
"loss": 2.7461,
|
| 1727 |
+
"step": 6050
|
| 1728 |
+
},
|
| 1729 |
+
{
|
| 1730 |
+
"epoch": 2.728109822857303,
|
| 1731 |
+
"grad_norm": 0.6948567628860474,
|
| 1732 |
+
"learning_rate": 0.0007496393785270148,
|
| 1733 |
+
"loss": 2.7297,
|
| 1734 |
+
"step": 6075
|
| 1735 |
+
},
|
| 1736 |
+
{
|
| 1737 |
+
"epoch": 2.739339153869908,
|
| 1738 |
+
"grad_norm": 0.5976940393447876,
|
| 1739 |
+
"learning_rate": 0.0007477045370256802,
|
| 1740 |
+
"loss": 2.7419,
|
| 1741 |
+
"step": 6100
|
| 1742 |
+
},
|
| 1743 |
+
{
|
| 1744 |
+
"epoch": 2.750568484882513,
|
| 1745 |
+
"grad_norm": 0.5582289099693298,
|
| 1746 |
+
"learning_rate": 0.0007457647672719133,
|
| 1747 |
+
"loss": 2.7238,
|
| 1748 |
+
"step": 6125
|
| 1749 |
+
},
|
| 1750 |
+
{
|
| 1751 |
+
"epoch": 2.761797815895118,
|
| 1752 |
+
"grad_norm": 0.6097133755683899,
|
| 1753 |
+
"learning_rate": 0.00074389798763174,
|
| 1754 |
+
"loss": 2.7345,
|
| 1755 |
+
"step": 6150
|
| 1756 |
+
},
|
| 1757 |
+
{
|
| 1758 |
+
"epoch": 2.7730271469077232,
|
| 1759 |
+
"grad_norm": 0.576604425907135,
|
| 1760 |
+
"learning_rate": 0.0007419486705442532,
|
| 1761 |
+
"loss": 2.7075,
|
| 1762 |
+
"step": 6175
|
| 1763 |
+
},
|
| 1764 |
+
{
|
| 1765 |
+
"epoch": 2.784256477920328,
|
| 1766 |
+
"grad_norm": 0.6071319580078125,
|
| 1767 |
+
"learning_rate": 0.0007399945397212636,
|
| 1768 |
+
"loss": 2.7122,
|
| 1769 |
+
"step": 6200
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"epoch": 2.795485808932933,
|
| 1773 |
+
"grad_norm": 0.6164671182632446,
|
| 1774 |
+
"learning_rate": 0.0007380356340415503,
|
| 1775 |
+
"loss": 2.698,
|
| 1776 |
+
"step": 6225
|
| 1777 |
+
},
|
| 1778 |
+
{
|
| 1779 |
+
"epoch": 2.806715139945538,
|
| 1780 |
+
"grad_norm": 0.5970659255981445,
|
| 1781 |
+
"learning_rate": 0.0007360719924788919,
|
| 1782 |
+
"loss": 2.7429,
|
| 1783 |
+
"step": 6250
|
| 1784 |
+
},
|
| 1785 |
+
{
|
| 1786 |
+
"epoch": 2.8179444709581425,
|
| 1787 |
+
"grad_norm": 0.5380481481552124,
|
| 1788 |
+
"learning_rate": 0.0007341036541012898,
|
| 1789 |
+
"loss": 2.6655,
|
| 1790 |
+
"step": 6275
|
| 1791 |
+
},
|
| 1792 |
+
{
|
| 1793 |
+
"epoch": 2.8291738019707475,
|
| 1794 |
+
"grad_norm": 0.5859966278076172,
|
| 1795 |
+
"learning_rate": 0.0007321306580701923,
|
| 1796 |
+
"loss": 2.7115,
|
| 1797 |
+
"step": 6300
|
| 1798 |
+
},
|
| 1799 |
+
{
|
| 1800 |
+
"epoch": 2.8404031329833526,
|
| 1801 |
+
"grad_norm": 0.5644718408584595,
|
| 1802 |
+
"learning_rate": 0.0007301530436397148,
|
| 1803 |
+
"loss": 2.6945,
|
| 1804 |
+
"step": 6325
|
| 1805 |
+
},
|
| 1806 |
+
{
|
| 1807 |
+
"epoch": 2.8516324639959576,
|
| 1808 |
+
"grad_norm": 0.5971605777740479,
|
| 1809 |
+
"learning_rate": 0.0007281708501558591,
|
| 1810 |
+
"loss": 2.7082,
|
| 1811 |
+
"step": 6350
|
| 1812 |
+
},
|
| 1813 |
+
{
|
| 1814 |
+
"epoch": 2.8628617950085626,
|
| 1815 |
+
"grad_norm": 0.6274561882019043,
|
| 1816 |
+
"learning_rate": 0.0007261841170557303,
|
| 1817 |
+
"loss": 2.7207,
|
| 1818 |
+
"step": 6375
|
| 1819 |
+
},
|
| 1820 |
+
{
|
| 1821 |
+
"epoch": 2.8740911260211672,
|
| 1822 |
+
"grad_norm": 0.611348032951355,
|
| 1823 |
+
"learning_rate": 0.0007241928838667522,
|
| 1824 |
+
"loss": 2.7155,
|
| 1825 |
+
"step": 6400
|
| 1826 |
+
},
|
| 1827 |
+
{
|
| 1828 |
+
"epoch": 2.8853204570337723,
|
| 1829 |
+
"grad_norm": 0.5650250911712646,
|
| 1830 |
+
"learning_rate": 0.000722197190205881,
|
| 1831 |
+
"loss": 2.7063,
|
| 1832 |
+
"step": 6425
|
| 1833 |
+
},
|
| 1834 |
+
{
|
| 1835 |
+
"epoch": 2.896549788046377,
|
| 1836 |
+
"grad_norm": 0.6043295860290527,
|
| 1837 |
+
"learning_rate": 0.0007201970757788173,
|
| 1838 |
+
"loss": 2.6909,
|
| 1839 |
+
"step": 6450
|
| 1840 |
+
},
|
| 1841 |
+
{
|
| 1842 |
+
"epoch": 2.907779119058982,
|
| 1843 |
+
"grad_norm": 0.582062304019928,
|
| 1844 |
+
"learning_rate": 0.0007181925803792153,
|
| 1845 |
+
"loss": 2.7262,
|
| 1846 |
+
"step": 6475
|
| 1847 |
+
},
|
| 1848 |
+
{
|
| 1849 |
+
"epoch": 2.919008450071587,
|
| 1850 |
+
"grad_norm": 0.6272075772285461,
|
| 1851 |
+
"learning_rate": 0.0007161837438878926,
|
| 1852 |
+
"loss": 2.7224,
|
| 1853 |
+
"step": 6500
|
| 1854 |
+
},
|
| 1855 |
+
{
|
| 1856 |
+
"epoch": 2.930237781084192,
|
| 1857 |
+
"grad_norm": 0.6399256587028503,
|
| 1858 |
+
"learning_rate": 0.0007141706062720349,
|
| 1859 |
+
"loss": 2.7202,
|
| 1860 |
+
"step": 6525
|
| 1861 |
+
},
|
| 1862 |
+
{
|
| 1863 |
+
"epoch": 2.941467112096797,
|
| 1864 |
+
"grad_norm": 0.637417197227478,
|
| 1865 |
+
"learning_rate": 0.0007121532075844023,
|
| 1866 |
+
"loss": 2.6624,
|
| 1867 |
+
"step": 6550
|
| 1868 |
+
},
|
| 1869 |
+
{
|
| 1870 |
+
"epoch": 2.9526964431094016,
|
| 1871 |
+
"grad_norm": 0.7301665544509888,
|
| 1872 |
+
"learning_rate": 0.0007101315879625315,
|
| 1873 |
+
"loss": 2.7103,
|
| 1874 |
+
"step": 6575
|
| 1875 |
+
},
|
| 1876 |
+
{
|
| 1877 |
+
"epoch": 2.9639257741220066,
|
| 1878 |
+
"grad_norm": 0.6282750964164734,
|
| 1879 |
+
"learning_rate": 0.000708105787627938,
|
| 1880 |
+
"loss": 2.6985,
|
| 1881 |
+
"step": 6600
|
| 1882 |
+
},
|
| 1883 |
+
{
|
| 1884 |
+
"epoch": 2.9751551051346117,
|
| 1885 |
+
"grad_norm": 0.7559499144554138,
|
| 1886 |
+
"learning_rate": 0.0007060758468853153,
|
| 1887 |
+
"loss": 2.6989,
|
| 1888 |
+
"step": 6625
|
| 1889 |
+
},
|
| 1890 |
+
{
|
| 1891 |
+
"epoch": 2.9863844361472163,
|
| 1892 |
+
"grad_norm": 0.6338511109352112,
|
| 1893 |
+
"learning_rate": 0.0007040418061217324,
|
| 1894 |
+
"loss": 2.7278,
|
| 1895 |
+
"step": 6650
|
| 1896 |
+
},
|
| 1897 |
+
{
|
| 1898 |
+
"epoch": 2.9976137671598213,
|
| 1899 |
+
"grad_norm": 0.6456329226493835,
|
| 1900 |
+
"learning_rate": 0.0007020037058058326,
|
| 1901 |
+
"loss": 2.6851,
|
| 1902 |
+
"step": 6675
|
| 1903 |
+
},
|
| 1904 |
+
{
|
| 1905 |
+
"epoch": 3.0,
|
| 1906 |
+
"eval_loss": 3.168698787689209,
|
| 1907 |
+
"eval_runtime": 228.8127,
|
| 1908 |
+
"eval_samples_per_second": 54.997,
|
| 1909 |
+
"eval_steps_per_second": 54.997,
|
| 1910 |
+
"step": 6681
|
| 1911 |
+
}
|
| 1912 |
+
],
|
| 1913 |
+
"logging_steps": 25,
|
| 1914 |
+
"max_steps": 17808,
|
| 1915 |
+
"num_input_tokens_seen": 0,
|
| 1916 |
+
"num_train_epochs": 8,
|
| 1917 |
+
"save_steps": 500,
|
| 1918 |
+
"stateful_callbacks": {
|
| 1919 |
+
"TrainerControl": {
|
| 1920 |
+
"args": {
|
| 1921 |
+
"should_epoch_stop": false,
|
| 1922 |
+
"should_evaluate": false,
|
| 1923 |
+
"should_log": false,
|
| 1924 |
+
"should_save": true,
|
| 1925 |
+
"should_training_stop": false
|
| 1926 |
+
},
|
| 1927 |
+
"attributes": {}
|
| 1928 |
+
}
|
| 1929 |
+
},
|
| 1930 |
+
"total_flos": 1.3044200408064e+16,
|
| 1931 |
+
"train_batch_size": 1,
|
| 1932 |
+
"trial_name": null,
|
| 1933 |
+
"trial_params": null
|
| 1934 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:489f044b9392a9ee28314cc59ac7352ed56662a402d578ab9e5c76a9b1cca731
|
| 3 |
+
size 5713
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|