Commit
•
55c8b65
1
Parent(s):
b9d48f4
Upload folder using huggingface_hub
Browse files- config.json +41 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +30 -0
- tokenizer.json +0 -0
- tokenizer_config.json +214 -0
- trainer_state.json +2391 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,41 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "EleutherAI/pythia-70m-deduped",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"GPTNeoXForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": true,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 0,
|
| 9 |
+
"classifier_dropout": 0.1,
|
| 10 |
+
"eos_token_id": 0,
|
| 11 |
+
"finetuning_task": "text-classification",
|
| 12 |
+
"hidden_act": "gelu",
|
| 13 |
+
"hidden_dropout": 0.0,
|
| 14 |
+
"hidden_size": 512,
|
| 15 |
+
"id2label": {
|
| 16 |
+
"0": "0",
|
| 17 |
+
"1": "1"
|
| 18 |
+
},
|
| 19 |
+
"initializer_range": 0.02,
|
| 20 |
+
"intermediate_size": 2048,
|
| 21 |
+
"label2id": {
|
| 22 |
+
"0": 0,
|
| 23 |
+
"1": 1
|
| 24 |
+
},
|
| 25 |
+
"layer_norm_eps": 1e-05,
|
| 26 |
+
"max_position_embeddings": 2048,
|
| 27 |
+
"model_type": "gpt_neox",
|
| 28 |
+
"num_attention_heads": 8,
|
| 29 |
+
"num_hidden_layers": 6,
|
| 30 |
+
"pad_token_id": 1,
|
| 31 |
+
"problem_type": "single_label_classification",
|
| 32 |
+
"rope_scaling": null,
|
| 33 |
+
"rotary_emb_base": 10000,
|
| 34 |
+
"rotary_pct": 0.25,
|
| 35 |
+
"tie_word_embeddings": false,
|
| 36 |
+
"torch_dtype": "float32",
|
| 37 |
+
"transformers_version": "4.40.0",
|
| 38 |
+
"use_cache": true,
|
| 39 |
+
"use_parallel_residual": true,
|
| 40 |
+
"vocab_size": 50304
|
| 41 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1118e7e407c295adf98a69a551aac9d6816cb611646788c343edd055ab66f75f
|
| 3 |
+
size 178696672
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a3a9a5936ab249c7ab9a75fcd851258c9cf4d44a5160702fae894b37b656d1ef
|
| 3 |
+
size 89894586
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:91879dfe5c74e96eb6cf7f48ae0d302d4303d02663cbf4f239b508d954e630d6
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b44ab142c282ef925b5fc1e59342c63b1ff0d373b002203b20538609ebd3ecc1
|
| 3 |
+
size 1064
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<|endoftext|>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "<|endoftext|>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": {
|
| 17 |
+
"content": "<|padding|>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"unk_token": {
|
| 24 |
+
"content": "<|endoftext|>",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
}
|
| 30 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,214 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": false,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"add_prefix_space": false,
|
| 5 |
+
"added_tokens_decoder": {
|
| 6 |
+
"0": {
|
| 7 |
+
"content": "<|endoftext|>",
|
| 8 |
+
"lstrip": false,
|
| 9 |
+
"normalized": false,
|
| 10 |
+
"rstrip": false,
|
| 11 |
+
"single_word": false,
|
| 12 |
+
"special": true
|
| 13 |
+
},
|
| 14 |
+
"1": {
|
| 15 |
+
"content": "<|padding|>",
|
| 16 |
+
"lstrip": false,
|
| 17 |
+
"normalized": false,
|
| 18 |
+
"rstrip": false,
|
| 19 |
+
"single_word": false,
|
| 20 |
+
"special": true
|
| 21 |
+
},
|
| 22 |
+
"50254": {
|
| 23 |
+
"content": " ",
|
| 24 |
+
"lstrip": false,
|
| 25 |
+
"normalized": true,
|
| 26 |
+
"rstrip": false,
|
| 27 |
+
"single_word": false,
|
| 28 |
+
"special": false
|
| 29 |
+
},
|
| 30 |
+
"50255": {
|
| 31 |
+
"content": " ",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": true,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false,
|
| 36 |
+
"special": false
|
| 37 |
+
},
|
| 38 |
+
"50256": {
|
| 39 |
+
"content": " ",
|
| 40 |
+
"lstrip": false,
|
| 41 |
+
"normalized": true,
|
| 42 |
+
"rstrip": false,
|
| 43 |
+
"single_word": false,
|
| 44 |
+
"special": false
|
| 45 |
+
},
|
| 46 |
+
"50257": {
|
| 47 |
+
"content": " ",
|
| 48 |
+
"lstrip": false,
|
| 49 |
+
"normalized": true,
|
| 50 |
+
"rstrip": false,
|
| 51 |
+
"single_word": false,
|
| 52 |
+
"special": false
|
| 53 |
+
},
|
| 54 |
+
"50258": {
|
| 55 |
+
"content": " ",
|
| 56 |
+
"lstrip": false,
|
| 57 |
+
"normalized": true,
|
| 58 |
+
"rstrip": false,
|
| 59 |
+
"single_word": false,
|
| 60 |
+
"special": false
|
| 61 |
+
},
|
| 62 |
+
"50259": {
|
| 63 |
+
"content": " ",
|
| 64 |
+
"lstrip": false,
|
| 65 |
+
"normalized": true,
|
| 66 |
+
"rstrip": false,
|
| 67 |
+
"single_word": false,
|
| 68 |
+
"special": false
|
| 69 |
+
},
|
| 70 |
+
"50260": {
|
| 71 |
+
"content": " ",
|
| 72 |
+
"lstrip": false,
|
| 73 |
+
"normalized": true,
|
| 74 |
+
"rstrip": false,
|
| 75 |
+
"single_word": false,
|
| 76 |
+
"special": false
|
| 77 |
+
},
|
| 78 |
+
"50261": {
|
| 79 |
+
"content": " ",
|
| 80 |
+
"lstrip": false,
|
| 81 |
+
"normalized": true,
|
| 82 |
+
"rstrip": false,
|
| 83 |
+
"single_word": false,
|
| 84 |
+
"special": false
|
| 85 |
+
},
|
| 86 |
+
"50262": {
|
| 87 |
+
"content": " ",
|
| 88 |
+
"lstrip": false,
|
| 89 |
+
"normalized": true,
|
| 90 |
+
"rstrip": false,
|
| 91 |
+
"single_word": false,
|
| 92 |
+
"special": false
|
| 93 |
+
},
|
| 94 |
+
"50263": {
|
| 95 |
+
"content": " ",
|
| 96 |
+
"lstrip": false,
|
| 97 |
+
"normalized": true,
|
| 98 |
+
"rstrip": false,
|
| 99 |
+
"single_word": false,
|
| 100 |
+
"special": false
|
| 101 |
+
},
|
| 102 |
+
"50264": {
|
| 103 |
+
"content": " ",
|
| 104 |
+
"lstrip": false,
|
| 105 |
+
"normalized": true,
|
| 106 |
+
"rstrip": false,
|
| 107 |
+
"single_word": false,
|
| 108 |
+
"special": false
|
| 109 |
+
},
|
| 110 |
+
"50265": {
|
| 111 |
+
"content": " ",
|
| 112 |
+
"lstrip": false,
|
| 113 |
+
"normalized": true,
|
| 114 |
+
"rstrip": false,
|
| 115 |
+
"single_word": false,
|
| 116 |
+
"special": false
|
| 117 |
+
},
|
| 118 |
+
"50266": {
|
| 119 |
+
"content": " ",
|
| 120 |
+
"lstrip": false,
|
| 121 |
+
"normalized": true,
|
| 122 |
+
"rstrip": false,
|
| 123 |
+
"single_word": false,
|
| 124 |
+
"special": false
|
| 125 |
+
},
|
| 126 |
+
"50267": {
|
| 127 |
+
"content": " ",
|
| 128 |
+
"lstrip": false,
|
| 129 |
+
"normalized": true,
|
| 130 |
+
"rstrip": false,
|
| 131 |
+
"single_word": false,
|
| 132 |
+
"special": false
|
| 133 |
+
},
|
| 134 |
+
"50268": {
|
| 135 |
+
"content": " ",
|
| 136 |
+
"lstrip": false,
|
| 137 |
+
"normalized": true,
|
| 138 |
+
"rstrip": false,
|
| 139 |
+
"single_word": false,
|
| 140 |
+
"special": false
|
| 141 |
+
},
|
| 142 |
+
"50269": {
|
| 143 |
+
"content": " ",
|
| 144 |
+
"lstrip": false,
|
| 145 |
+
"normalized": true,
|
| 146 |
+
"rstrip": false,
|
| 147 |
+
"single_word": false,
|
| 148 |
+
"special": false
|
| 149 |
+
},
|
| 150 |
+
"50270": {
|
| 151 |
+
"content": " ",
|
| 152 |
+
"lstrip": false,
|
| 153 |
+
"normalized": true,
|
| 154 |
+
"rstrip": false,
|
| 155 |
+
"single_word": false,
|
| 156 |
+
"special": false
|
| 157 |
+
},
|
| 158 |
+
"50271": {
|
| 159 |
+
"content": " ",
|
| 160 |
+
"lstrip": false,
|
| 161 |
+
"normalized": true,
|
| 162 |
+
"rstrip": false,
|
| 163 |
+
"single_word": false,
|
| 164 |
+
"special": false
|
| 165 |
+
},
|
| 166 |
+
"50272": {
|
| 167 |
+
"content": " ",
|
| 168 |
+
"lstrip": false,
|
| 169 |
+
"normalized": true,
|
| 170 |
+
"rstrip": false,
|
| 171 |
+
"single_word": false,
|
| 172 |
+
"special": false
|
| 173 |
+
},
|
| 174 |
+
"50273": {
|
| 175 |
+
"content": " ",
|
| 176 |
+
"lstrip": false,
|
| 177 |
+
"normalized": true,
|
| 178 |
+
"rstrip": false,
|
| 179 |
+
"single_word": false,
|
| 180 |
+
"special": false
|
| 181 |
+
},
|
| 182 |
+
"50274": {
|
| 183 |
+
"content": " ",
|
| 184 |
+
"lstrip": false,
|
| 185 |
+
"normalized": true,
|
| 186 |
+
"rstrip": false,
|
| 187 |
+
"single_word": false,
|
| 188 |
+
"special": false
|
| 189 |
+
},
|
| 190 |
+
"50275": {
|
| 191 |
+
"content": " ",
|
| 192 |
+
"lstrip": false,
|
| 193 |
+
"normalized": true,
|
| 194 |
+
"rstrip": false,
|
| 195 |
+
"single_word": false,
|
| 196 |
+
"special": false
|
| 197 |
+
},
|
| 198 |
+
"50276": {
|
| 199 |
+
"content": " ",
|
| 200 |
+
"lstrip": false,
|
| 201 |
+
"normalized": true,
|
| 202 |
+
"rstrip": false,
|
| 203 |
+
"single_word": false,
|
| 204 |
+
"special": false
|
| 205 |
+
}
|
| 206 |
+
},
|
| 207 |
+
"bos_token": "<|endoftext|>",
|
| 208 |
+
"clean_up_tokenization_spaces": true,
|
| 209 |
+
"eos_token": "<|endoftext|>",
|
| 210 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 211 |
+
"pad_token": "<|padding|>",
|
| 212 |
+
"tokenizer_class": "GPTNeoXTokenizer",
|
| 213 |
+
"unk_token": "<|endoftext|>"
|
| 214 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2391 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.4821594787465093,
|
| 5 |
+
"eval_steps": 50,
|
| 6 |
+
"global_step": 1500,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.008273864929155031,
|
| 13 |
+
"grad_norm": 193.22154235839844,
|
| 14 |
+
"learning_rate": 8.998478317951598e-06,
|
| 15 |
+
"loss": 1.2655,
|
| 16 |
+
"step": 5
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.016547729858310063,
|
| 20 |
+
"grad_norm": 224.26678466796875,
|
| 21 |
+
"learning_rate": 8.993914300924725e-06,
|
| 22 |
+
"loss": 0.8096,
|
| 23 |
+
"step": 10
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.024821594787465096,
|
| 27 |
+
"grad_norm": 233.3835906982422,
|
| 28 |
+
"learning_rate": 8.986311035578395e-06,
|
| 29 |
+
"loss": 0.7882,
|
| 30 |
+
"step": 15
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.033095459716620125,
|
| 34 |
+
"grad_norm": 118.54664611816406,
|
| 35 |
+
"learning_rate": 8.97567366402478e-06,
|
| 36 |
+
"loss": 0.642,
|
| 37 |
+
"step": 20
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.041369324645775155,
|
| 41 |
+
"grad_norm": 156.66001892089844,
|
| 42 |
+
"learning_rate": 8.962009380351584e-06,
|
| 43 |
+
"loss": 0.7024,
|
| 44 |
+
"step": 25
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.04964318957493019,
|
| 48 |
+
"grad_norm": 272.93414306640625,
|
| 49 |
+
"learning_rate": 8.945327425756662e-06,
|
| 50 |
+
"loss": 0.6812,
|
| 51 |
+
"step": 30
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.05791705450408522,
|
| 55 |
+
"grad_norm": 197.8726043701172,
|
| 56 |
+
"learning_rate": 8.925639082298161e-06,
|
| 57 |
+
"loss": 0.7415,
|
| 58 |
+
"step": 35
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.06619091943324025,
|
| 62 |
+
"grad_norm": 114.77101135253906,
|
| 63 |
+
"learning_rate": 8.902957665264444e-06,
|
| 64 |
+
"loss": 0.6585,
|
| 65 |
+
"step": 40
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.07446478436239529,
|
| 69 |
+
"grad_norm": 104.65145874023438,
|
| 70 |
+
"learning_rate": 8.8772985141689e-06,
|
| 71 |
+
"loss": 0.675,
|
| 72 |
+
"step": 45
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.08273864929155031,
|
| 76 |
+
"grad_norm": 115.49458312988281,
|
| 77 |
+
"learning_rate": 8.848678982375792e-06,
|
| 78 |
+
"loss": 0.6154,
|
| 79 |
+
"step": 50
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.08273864929155031,
|
| 83 |
+
"eval_accuracy": 0.6920930232558139,
|
| 84 |
+
"eval_loss": 0.6675890684127808,
|
| 85 |
+
"eval_runtime": 135.1534,
|
| 86 |
+
"eval_samples_per_second": 7.954,
|
| 87 |
+
"eval_steps_per_second": 7.954,
|
| 88 |
+
"step": 50
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 0.09101251422070535,
|
| 92 |
+
"grad_norm": 236.3074951171875,
|
| 93 |
+
"learning_rate": 8.817118425364132e-06,
|
| 94 |
+
"loss": 0.627,
|
| 95 |
+
"step": 55
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 0.09928637914986038,
|
| 99 |
+
"grad_norm": 75.98784637451172,
|
| 100 |
+
"learning_rate": 8.78263818763749e-06,
|
| 101 |
+
"loss": 0.5869,
|
| 102 |
+
"step": 60
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 0.1075602440790154,
|
| 106 |
+
"grad_norm": 216.54205322265625,
|
| 107 |
+
"learning_rate": 8.745261588288655e-06,
|
| 108 |
+
"loss": 0.6362,
|
| 109 |
+
"step": 65
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.11583410900817044,
|
| 113 |
+
"grad_norm": 145.553466796875,
|
| 114 |
+
"learning_rate": 8.705013905228855e-06,
|
| 115 |
+
"loss": 0.7391,
|
| 116 |
+
"step": 70
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.12410797393732548,
|
| 120 |
+
"grad_norm": 62.87096405029297,
|
| 121 |
+
"learning_rate": 8.661922358092222e-06,
|
| 122 |
+
"loss": 0.6121,
|
| 123 |
+
"step": 75
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.1323818388664805,
|
| 127 |
+
"grad_norm": 50.98234176635742,
|
| 128 |
+
"learning_rate": 8.616016089827077e-06,
|
| 129 |
+
"loss": 0.6115,
|
| 130 |
+
"step": 80
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.14065570379563552,
|
| 134 |
+
"grad_norm": 63.97044372558594,
|
| 135 |
+
"learning_rate": 8.567326146986454e-06,
|
| 136 |
+
"loss": 0.6572,
|
| 137 |
+
"step": 85
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.14892956872479057,
|
| 141 |
+
"grad_norm": 101.70757293701172,
|
| 142 |
+
"learning_rate": 8.51588545873122e-06,
|
| 143 |
+
"loss": 0.8113,
|
| 144 |
+
"step": 90
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.1572034336539456,
|
| 148 |
+
"grad_norm": 174.54574584960938,
|
| 149 |
+
"learning_rate": 8.461728814559992e-06,
|
| 150 |
+
"loss": 0.7307,
|
| 151 |
+
"step": 95
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.16547729858310062,
|
| 155 |
+
"grad_norm": 179.63229370117188,
|
| 156 |
+
"learning_rate": 8.404892840780868e-06,
|
| 157 |
+
"loss": 0.6292,
|
| 158 |
+
"step": 100
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.16547729858310062,
|
| 162 |
+
"eval_accuracy": 0.6846511627906977,
|
| 163 |
+
"eval_loss": 0.7192373871803284,
|
| 164 |
+
"eval_runtime": 136.6349,
|
| 165 |
+
"eval_samples_per_second": 7.868,
|
| 166 |
+
"eval_steps_per_second": 7.868,
|
| 167 |
+
"step": 100
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 0.17375116351225567,
|
| 171 |
+
"grad_norm": 177.11233520507812,
|
| 172 |
+
"learning_rate": 8.345415975740966e-06,
|
| 173 |
+
"loss": 0.853,
|
| 174 |
+
"step": 105
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.1820250284414107,
|
| 178 |
+
"grad_norm": 83.83477020263672,
|
| 179 |
+
"learning_rate": 8.283338443830433e-06,
|
| 180 |
+
"loss": 0.667,
|
| 181 |
+
"step": 110
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.1902988933705657,
|
| 185 |
+
"grad_norm": 65.83185577392578,
|
| 186 |
+
"learning_rate": 8.21870222827856e-06,
|
| 187 |
+
"loss": 0.7123,
|
| 188 |
+
"step": 115
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.19857275829972076,
|
| 192 |
+
"grad_norm": 99.53777313232422,
|
| 193 |
+
"learning_rate": 8.151551042760408e-06,
|
| 194 |
+
"loss": 0.6869,
|
| 195 |
+
"step": 120
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 0.2068466232288758,
|
| 199 |
+
"grad_norm": 55.019622802734375,
|
| 200 |
+
"learning_rate": 8.081930301833101e-06,
|
| 201 |
+
"loss": 0.6729,
|
| 202 |
+
"step": 125
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 0.2151204881580308,
|
| 206 |
+
"grad_norm": 99.40830993652344,
|
| 207 |
+
"learning_rate": 8.009887090221828e-06,
|
| 208 |
+
"loss": 0.6269,
|
| 209 |
+
"step": 130
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.22339435308718586,
|
| 213 |
+
"grad_norm": 30.939523696899414,
|
| 214 |
+
"learning_rate": 7.935470130976281e-06,
|
| 215 |
+
"loss": 0.5331,
|
| 216 |
+
"step": 135
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 0.23166821801634088,
|
| 220 |
+
"grad_norm": 195.37283325195312,
|
| 221 |
+
"learning_rate": 7.858729752519121e-06,
|
| 222 |
+
"loss": 0.6496,
|
| 223 |
+
"step": 140
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 0.2399420829454959,
|
| 227 |
+
"grad_norm": 163.1190185546875,
|
| 228 |
+
"learning_rate": 7.779717854608697e-06,
|
| 229 |
+
"loss": 0.6195,
|
| 230 |
+
"step": 145
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.24821594787465096,
|
| 234 |
+
"grad_norm": 142.21194458007812,
|
| 235 |
+
"learning_rate": 7.69848787323908e-06,
|
| 236 |
+
"loss": 0.6002,
|
| 237 |
+
"step": 150
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 0.24821594787465096,
|
| 241 |
+
"eval_accuracy": 0.6781395348837209,
|
| 242 |
+
"eval_loss": 0.6401144862174988,
|
| 243 |
+
"eval_runtime": 135.4948,
|
| 244 |
+
"eval_samples_per_second": 7.934,
|
| 245 |
+
"eval_steps_per_second": 7.934,
|
| 246 |
+
"step": 150
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 0.256489812803806,
|
| 250 |
+
"grad_norm": 33.41708755493164,
|
| 251 |
+
"learning_rate": 7.615094744501132e-06,
|
| 252 |
+
"loss": 0.6123,
|
| 253 |
+
"step": 155
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 0.264763677732961,
|
| 257 |
+
"grad_norm": 134.9268798828125,
|
| 258 |
+
"learning_rate": 7.5295948674290595e-06,
|
| 259 |
+
"loss": 0.6182,
|
| 260 |
+
"step": 160
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.27303754266211605,
|
| 264 |
+
"grad_norm": 161.4491729736328,
|
| 265 |
+
"learning_rate": 7.442046065857564e-06,
|
| 266 |
+
"loss": 0.6316,
|
| 267 |
+
"step": 165
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 0.28131140759127105,
|
| 271 |
+
"grad_norm": 107.71475219726562,
|
| 272 |
+
"learning_rate": 7.352507549315408e-06,
|
| 273 |
+
"loss": 0.6307,
|
| 274 |
+
"step": 170
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 0.2895852725204261,
|
| 278 |
+
"grad_norm": 51.73357391357422,
|
| 279 |
+
"learning_rate": 7.261039872981816e-06,
|
| 280 |
+
"loss": 0.556,
|
| 281 |
+
"step": 175
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 0.29785913744958115,
|
| 285 |
+
"grad_norm": 76.03494262695312,
|
| 286 |
+
"learning_rate": 7.167704896732828e-06,
|
| 287 |
+
"loss": 0.7206,
|
| 288 |
+
"step": 180
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 0.30613300237873614,
|
| 292 |
+
"grad_norm": 62.3538818359375,
|
| 293 |
+
"learning_rate": 7.0725657433052645e-06,
|
| 294 |
+
"loss": 0.5621,
|
| 295 |
+
"step": 185
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 0.3144068673078912,
|
| 299 |
+
"grad_norm": 61.0316047668457,
|
| 300 |
+
"learning_rate": 6.975686755606624e-06,
|
| 301 |
+
"loss": 0.6524,
|
| 302 |
+
"step": 190
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.32268073223704624,
|
| 306 |
+
"grad_norm": 29.230974197387695,
|
| 307 |
+
"learning_rate": 6.877133453199773e-06,
|
| 308 |
+
"loss": 0.5413,
|
| 309 |
+
"step": 195
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.33095459716620124,
|
| 313 |
+
"grad_norm": 137.9040069580078,
|
| 314 |
+
"learning_rate": 6.7769724879918564e-06,
|
| 315 |
+
"loss": 0.6446,
|
| 316 |
+
"step": 200
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 0.33095459716620124,
|
| 320 |
+
"eval_accuracy": 0.6837209302325581,
|
| 321 |
+
"eval_loss": 0.6216422319412231,
|
| 322 |
+
"eval_runtime": 134.8141,
|
| 323 |
+
"eval_samples_per_second": 7.974,
|
| 324 |
+
"eval_steps_per_second": 7.974,
|
| 325 |
+
"step": 200
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.3392284620953563,
|
| 329 |
+
"grad_norm": 267.3862609863281,
|
| 330 |
+
"learning_rate": 6.675271599157415e-06,
|
| 331 |
+
"loss": 0.6703,
|
| 332 |
+
"step": 205
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.34750232702451134,
|
| 336 |
+
"grad_norm": 50.05187225341797,
|
| 337 |
+
"learning_rate": 6.572099567326157e-06,
|
| 338 |
+
"loss": 0.603,
|
| 339 |
+
"step": 210
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.35577619195366633,
|
| 343 |
+
"grad_norm": 102.92433166503906,
|
| 344 |
+
"learning_rate": 6.467526168066408e-06,
|
| 345 |
+
"loss": 0.6523,
|
| 346 |
+
"step": 215
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.3640500568828214,
|
| 350 |
+
"grad_norm": 114.91250610351562,
|
| 351 |
+
"learning_rate": 6.361622124695677e-06,
|
| 352 |
+
"loss": 0.5757,
|
| 353 |
+
"step": 220
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.37232392181197643,
|
| 357 |
+
"grad_norm": 48.830989837646484,
|
| 358 |
+
"learning_rate": 6.254459060450252e-06,
|
| 359 |
+
"loss": 0.6294,
|
| 360 |
+
"step": 225
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.3805977867411314,
|
| 364 |
+
"grad_norm": 36.482215881347656,
|
| 365 |
+
"learning_rate": 6.146109450046187e-06,
|
| 366 |
+
"loss": 0.5982,
|
| 367 |
+
"step": 230
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.3888716516702865,
|
| 371 |
+
"grad_norm": 38.25822830200195,
|
| 372 |
+
"learning_rate": 6.036646570664412e-06,
|
| 373 |
+
"loss": 0.5573,
|
| 374 |
+
"step": 235
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.39714551659944153,
|
| 378 |
+
"grad_norm": 73.86414337158203,
|
| 379 |
+
"learning_rate": 5.926144452393163e-06,
|
| 380 |
+
"loss": 0.6372,
|
| 381 |
+
"step": 240
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.4054193815285965,
|
| 385 |
+
"grad_norm": 34.50651550292969,
|
| 386 |
+
"learning_rate": 5.814677828161186e-06,
|
| 387 |
+
"loss": 0.5968,
|
| 388 |
+
"step": 245
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.4136932464577516,
|
| 392 |
+
"grad_norm": 109.74346160888672,
|
| 393 |
+
"learning_rate": 5.7023220831956335e-06,
|
| 394 |
+
"loss": 0.673,
|
| 395 |
+
"step": 250
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.4136932464577516,
|
| 399 |
+
"eval_accuracy": 0.6874418604651162,
|
| 400 |
+
"eval_loss": 0.6246343851089478,
|
| 401 |
+
"eval_runtime": 136.0439,
|
| 402 |
+
"eval_samples_per_second": 7.902,
|
| 403 |
+
"eval_steps_per_second": 7.902,
|
| 404 |
+
"step": 250
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 0.4219671113869066,
|
| 408 |
+
"grad_norm": 91.58174133300781,
|
| 409 |
+
"learning_rate": 5.589153204038793e-06,
|
| 410 |
+
"loss": 0.5698,
|
| 411 |
+
"step": 255
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 0.4302409763160616,
|
| 415 |
+
"grad_norm": 162.50047302246094,
|
| 416 |
+
"learning_rate": 5.475247727158154e-06,
|
| 417 |
+
"loss": 0.6554,
|
| 418 |
+
"step": 260
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 0.43851484124521667,
|
| 422 |
+
"grad_norm": 144.00473022460938,
|
| 423 |
+
"learning_rate": 5.360682687184554e-06,
|
| 424 |
+
"loss": 0.5838,
|
| 425 |
+
"step": 265
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 0.4467887061743717,
|
| 429 |
+
"grad_norm": 94.68612670898438,
|
| 430 |
+
"learning_rate": 5.2455355648134174e-06,
|
| 431 |
+
"loss": 0.5319,
|
| 432 |
+
"step": 270
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 0.4550625711035267,
|
| 436 |
+
"grad_norm": 55.78764724731445,
|
| 437 |
+
"learning_rate": 5.129884234404314e-06,
|
| 438 |
+
"loss": 0.5505,
|
| 439 |
+
"step": 275
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 0.46333643603268176,
|
| 443 |
+
"grad_norm": 132.5346221923828,
|
| 444 |
+
"learning_rate": 5.013806911314294e-06,
|
| 445 |
+
"loss": 0.6897,
|
| 446 |
+
"step": 280
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 0.4716103009618368,
|
| 450 |
+
"grad_norm": 114.60745239257812,
|
| 451 |
+
"learning_rate": 4.897382099000587e-06,
|
| 452 |
+
"loss": 0.6442,
|
| 453 |
+
"step": 285
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 0.4798841658909918,
|
| 457 |
+
"grad_norm": 35.80769348144531,
|
| 458 |
+
"learning_rate": 4.780688535928477e-06,
|
| 459 |
+
"loss": 0.7135,
|
| 460 |
+
"step": 290
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 0.48815803082014686,
|
| 464 |
+
"grad_norm": 102.89435577392578,
|
| 465 |
+
"learning_rate": 4.66380514232023e-06,
|
| 466 |
+
"loss": 0.605,
|
| 467 |
+
"step": 295
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 0.4964318957493019,
|
| 471 |
+
"grad_norm": 34.77372741699219,
|
| 472 |
+
"learning_rate": 4.546810966781089e-06,
|
| 473 |
+
"loss": 0.711,
|
| 474 |
+
"step": 300
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 0.4964318957493019,
|
| 478 |
+
"eval_accuracy": 0.6827906976744186,
|
| 479 |
+
"eval_loss": 0.6517416834831238,
|
| 480 |
+
"eval_runtime": 135.0365,
|
| 481 |
+
"eval_samples_per_second": 7.961,
|
| 482 |
+
"eval_steps_per_second": 7.961,
|
| 483 |
+
"step": 300
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 0.504705760678457,
|
| 487 |
+
"grad_norm": 88.30133819580078,
|
| 488 |
+
"learning_rate": 4.429785132838475e-06,
|
| 489 |
+
"loss": 0.5888,
|
| 490 |
+
"step": 305
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 0.512979625607612,
|
| 494 |
+
"grad_norm": 29.411426544189453,
|
| 495 |
+
"learning_rate": 4.312806785430477e-06,
|
| 496 |
+
"loss": 0.566,
|
| 497 |
+
"step": 310
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 0.521253490536767,
|
| 501 |
+
"grad_norm": 118.55986022949219,
|
| 502 |
+
"learning_rate": 4.195955037379899e-06,
|
| 503 |
+
"loss": 0.5398,
|
| 504 |
+
"step": 315
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 0.529527355465922,
|
| 508 |
+
"grad_norm": 57.91450500488281,
|
| 509 |
+
"learning_rate": 4.079308915889999e-06,
|
| 510 |
+
"loss": 0.5738,
|
| 511 |
+
"step": 320
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 0.537801220395077,
|
| 515 |
+
"grad_norm": 61.48877716064453,
|
| 516 |
+
"learning_rate": 3.96294730909815e-06,
|
| 517 |
+
"loss": 0.5945,
|
| 518 |
+
"step": 325
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 0.5460750853242321,
|
| 522 |
+
"grad_norm": 48.938419342041016,
|
| 523 |
+
"learning_rate": 3.846948912723542e-06,
|
| 524 |
+
"loss": 0.6338,
|
| 525 |
+
"step": 330
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 0.5543489502533872,
|
| 529 |
+
"grad_norm": 145.6968231201172,
|
| 530 |
+
"learning_rate": 3.731392176845023e-06,
|
| 531 |
+
"loss": 0.5681,
|
| 532 |
+
"step": 335
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 0.5626228151825421,
|
| 536 |
+
"grad_norm": 105.9911117553711,
|
| 537 |
+
"learning_rate": 3.6163552528450616e-06,
|
| 538 |
+
"loss": 0.6693,
|
| 539 |
+
"step": 340
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 0.5708966801116971,
|
| 543 |
+
"grad_norm": 140.98574829101562,
|
| 544 |
+
"learning_rate": 3.501915940555728e-06,
|
| 545 |
+
"loss": 0.6621,
|
| 546 |
+
"step": 345
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 0.5791705450408522,
|
| 550 |
+
"grad_norm": 92.60147094726562,
|
| 551 |
+
"learning_rate": 3.3881516356424178e-06,
|
| 552 |
+
"loss": 0.6682,
|
| 553 |
+
"step": 350
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 0.5791705450408522,
|
| 557 |
+
"eval_accuracy": 0.6809302325581396,
|
| 558 |
+
"eval_loss": 0.6307579278945923,
|
| 559 |
+
"eval_runtime": 136.0513,
|
| 560 |
+
"eval_samples_per_second": 7.901,
|
| 561 |
+
"eval_steps_per_second": 7.901,
|
| 562 |
+
"step": 350
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.5874444099700072,
|
| 566 |
+
"grad_norm": 121.18115997314453,
|
| 567 |
+
"learning_rate": 3.275139277260925e-06,
|
| 568 |
+
"loss": 0.677,
|
| 569 |
+
"step": 355
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.5957182748991623,
|
| 573 |
+
"grad_norm": 158.01596069335938,
|
| 574 |
+
"learning_rate": 3.162955296023247e-06,
|
| 575 |
+
"loss": 0.6703,
|
| 576 |
+
"step": 360
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.6039921398283173,
|
| 580 |
+
"grad_norm": 59.304481506347656,
|
| 581 |
+
"learning_rate": 3.0516755623073273e-06,
|
| 582 |
+
"loss": 0.5698,
|
| 583 |
+
"step": 365
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.6122660047574723,
|
| 587 |
+
"grad_norm": 40.63115310668945,
|
| 588 |
+
"learning_rate": 2.941375334945674e-06,
|
| 589 |
+
"loss": 0.5812,
|
| 590 |
+
"step": 370
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.6205398696866273,
|
| 594 |
+
"grad_norm": 71.80590057373047,
|
| 595 |
+
"learning_rate": 2.8321292103275813e-06,
|
| 596 |
+
"loss": 0.5766,
|
| 597 |
+
"step": 375
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.6288137346157824,
|
| 601 |
+
"grad_norm": 22.09684944152832,
|
| 602 |
+
"learning_rate": 2.7240110719493568e-06,
|
| 603 |
+
"loss": 0.5898,
|
| 604 |
+
"step": 380
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.6370875995449374,
|
| 608 |
+
"grad_norm": 32.82063293457031,
|
| 609 |
+
"learning_rate": 2.617094040446676e-06,
|
| 610 |
+
"loss": 0.6404,
|
| 611 |
+
"step": 385
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.6453614644740925,
|
| 615 |
+
"grad_norm": 35.220211029052734,
|
| 616 |
+
"learning_rate": 2.511450424142878e-06,
|
| 617 |
+
"loss": 0.649,
|
| 618 |
+
"step": 390
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.6536353294032475,
|
| 622 |
+
"grad_norm": 22.362548828125,
|
| 623 |
+
"learning_rate": 2.4071516701466072e-06,
|
| 624 |
+
"loss": 0.5896,
|
| 625 |
+
"step": 395
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.6619091943324025,
|
| 629 |
+
"grad_norm": 159.33010864257812,
|
| 630 |
+
"learning_rate": 2.304268316031922e-06,
|
| 631 |
+
"loss": 0.5437,
|
| 632 |
+
"step": 400
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.6619091943324025,
|
| 636 |
+
"eval_accuracy": 0.68,
|
| 637 |
+
"eval_loss": 0.6238510608673096,
|
| 638 |
+
"eval_runtime": 137.5204,
|
| 639 |
+
"eval_samples_per_second": 7.817,
|
| 640 |
+
"eval_steps_per_second": 7.817,
|
| 641 |
+
"step": 400
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 0.6701830592615575,
|
| 645 |
+
"grad_norm": 15.642170906066895,
|
| 646 |
+
"learning_rate": 2.2028699421335074e-06,
|
| 647 |
+
"loss": 0.5992,
|
| 648 |
+
"step": 405
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 0.6784569241907126,
|
| 652 |
+
"grad_norm": 84.97122955322266,
|
| 653 |
+
"learning_rate": 2.1030251244892713e-06,
|
| 654 |
+
"loss": 0.6152,
|
| 655 |
+
"step": 410
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 0.6867307891198676,
|
| 659 |
+
"grad_norm": 22.397871017456055,
|
| 660 |
+
"learning_rate": 2.0048013884621617e-06,
|
| 661 |
+
"loss": 0.5848,
|
| 662 |
+
"step": 415
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 0.6950046540490227,
|
| 666 |
+
"grad_norm": 102.53570556640625,
|
| 667 |
+
"learning_rate": 1.908265163072554e-06,
|
| 668 |
+
"loss": 0.5884,
|
| 669 |
+
"step": 420
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 0.7032785189781777,
|
| 673 |
+
"grad_norm": 175.41653442382812,
|
| 674 |
+
"learning_rate": 1.81348173607209e-06,
|
| 675 |
+
"loss": 0.575,
|
| 676 |
+
"step": 425
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 0.7115523839073327,
|
| 680 |
+
"grad_norm": 50.430362701416016,
|
| 681 |
+
"learning_rate": 1.7205152097893694e-06,
|
| 682 |
+
"loss": 0.6514,
|
| 683 |
+
"step": 430
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 0.7198262488364877,
|
| 687 |
+
"grad_norm": 121.63434600830078,
|
| 688 |
+
"learning_rate": 1.6294284577773493e-06,
|
| 689 |
+
"loss": 0.5527,
|
| 690 |
+
"step": 435
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 0.7281001137656428,
|
| 694 |
+
"grad_norm": 29.108440399169922,
|
| 695 |
+
"learning_rate": 1.540283082291754e-06,
|
| 696 |
+
"loss": 0.6797,
|
| 697 |
+
"step": 440
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 0.7363739786947978,
|
| 701 |
+
"grad_norm": 71.40788269042969,
|
| 702 |
+
"learning_rate": 1.4531393726292826e-06,
|
| 703 |
+
"loss": 0.6111,
|
| 704 |
+
"step": 445
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 0.7446478436239529,
|
| 708 |
+
"grad_norm": 43.776885986328125,
|
| 709 |
+
"learning_rate": 1.3680562643537693e-06,
|
| 710 |
+
"loss": 0.6352,
|
| 711 |
+
"step": 450
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 0.7446478436239529,
|
| 715 |
+
"eval_accuracy": 0.6827906976744186,
|
| 716 |
+
"eval_loss": 0.6232383251190186,
|
| 717 |
+
"eval_runtime": 142.7642,
|
| 718 |
+
"eval_samples_per_second": 7.53,
|
| 719 |
+
"eval_steps_per_second": 7.53,
|
| 720 |
+
"step": 450
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 0.7529217085531079,
|
| 724 |
+
"grad_norm": 169.23094177246094,
|
| 725 |
+
"learning_rate": 1.285091299437875e-06,
|
| 726 |
+
"loss": 0.537,
|
| 727 |
+
"step": 455
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 0.7611955734822629,
|
| 731 |
+
"grad_norm": 154.0270538330078,
|
| 732 |
+
"learning_rate": 1.2043005873472697e-06,
|
| 733 |
+
"loss": 0.5469,
|
| 734 |
+
"step": 460
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 0.7694694384114179,
|
| 738 |
+
"grad_norm": 106.58590698242188,
|
| 739 |
+
"learning_rate": 1.125738767093626e-06,
|
| 740 |
+
"loss": 0.6973,
|
| 741 |
+
"step": 465
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 0.777743303340573,
|
| 745 |
+
"grad_norm": 52.2432746887207,
|
| 746 |
+
"learning_rate": 1.049458970282088e-06,
|
| 747 |
+
"loss": 0.6092,
|
| 748 |
+
"step": 470
|
| 749 |
+
},
|
| 750 |
+
{
|
| 751 |
+
"epoch": 0.786017168269728,
|
| 752 |
+
"grad_norm": 27.52301788330078,
|
| 753 |
+
"learning_rate": 9.755127851781945e-07,
|
| 754 |
+
"loss": 0.5584,
|
| 755 |
+
"step": 475
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 0.7942910331988831,
|
| 759 |
+
"grad_norm": 113.73265075683594,
|
| 760 |
+
"learning_rate": 9.039502218185748e-07,
|
| 761 |
+
"loss": 0.5841,
|
| 762 |
+
"step": 480
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 0.8025648981280381,
|
| 766 |
+
"grad_norm": 85.42076110839844,
|
| 767 |
+
"learning_rate": 8.348196781890096e-07,
|
| 768 |
+
"loss": 0.6151,
|
| 769 |
+
"step": 485
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 0.810838763057193,
|
| 773 |
+
"grad_norm": 40.63630294799805,
|
| 774 |
+
"learning_rate": 7.681679074927166e-07,
|
| 775 |
+
"loss": 0.5904,
|
| 776 |
+
"step": 490
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 0.8191126279863481,
|
| 780 |
+
"grad_norm": 50.79077911376953,
|
| 781 |
+
"learning_rate": 7.04039986531011e-07,
|
| 782 |
+
"loss": 0.6572,
|
| 783 |
+
"step": 495
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 0.8273864929155031,
|
| 787 |
+
"grad_norm": 53.14852523803711,
|
| 788 |
+
"learning_rate": 6.424792852177275e-07,
|
| 789 |
+
"loss": 0.6784,
|
| 790 |
+
"step": 500
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 0.8273864929155031,
|
| 794 |
+
"eval_accuracy": 0.6865116279069767,
|
| 795 |
+
"eval_loss": 0.621077835559845,
|
| 796 |
+
"eval_runtime": 142.2156,
|
| 797 |
+
"eval_samples_per_second": 7.559,
|
| 798 |
+
"eval_steps_per_second": 7.559,
|
| 799 |
+
"step": 500
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 0.8356603578446582,
|
| 803 |
+
"grad_norm": 240.6154022216797,
|
| 804 |
+
"learning_rate": 7.382571090205965e-06,
|
| 805 |
+
"loss": 0.6728,
|
| 806 |
+
"step": 505
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 0.8439342227738132,
|
| 810 |
+
"grad_norm": 38.96548843383789,
|
| 811 |
+
"learning_rate": 7.352507549315408e-06,
|
| 812 |
+
"loss": 0.6614,
|
| 813 |
+
"step": 510
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 0.8522080877029683,
|
| 817 |
+
"grad_norm": 20.97259521484375,
|
| 818 |
+
"learning_rate": 7.322229646597724e-06,
|
| 819 |
+
"loss": 0.5843,
|
| 820 |
+
"step": 515
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 0.8604819526321232,
|
| 824 |
+
"grad_norm": 108.45309448242188,
|
| 825 |
+
"learning_rate": 7.2917396573936266e-06,
|
| 826 |
+
"loss": 0.6245,
|
| 827 |
+
"step": 520
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 0.8687558175612783,
|
| 831 |
+
"grad_norm": 142.10877990722656,
|
| 832 |
+
"learning_rate": 7.261039872981816e-06,
|
| 833 |
+
"loss": 0.5639,
|
| 834 |
+
"step": 525
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 0.8770296824904333,
|
| 838 |
+
"grad_norm": 74.65398406982422,
|
| 839 |
+
"learning_rate": 7.2301326004068e-06,
|
| 840 |
+
"loss": 0.5823,
|
| 841 |
+
"step": 530
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 0.8853035474195884,
|
| 845 |
+
"grad_norm": 89.85851287841797,
|
| 846 |
+
"learning_rate": 7.199020162305524e-06,
|
| 847 |
+
"loss": 0.5769,
|
| 848 |
+
"step": 535
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 0.8935774123487434,
|
| 852 |
+
"grad_norm": 60.187286376953125,
|
| 853 |
+
"learning_rate": 7.167704896732828e-06,
|
| 854 |
+
"loss": 0.6494,
|
| 855 |
+
"step": 540
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 0.9018512772778985,
|
| 859 |
+
"grad_norm": 77.58560180664062,
|
| 860 |
+
"learning_rate": 7.136189156985742e-06,
|
| 861 |
+
"loss": 0.6549,
|
| 862 |
+
"step": 545
|
| 863 |
+
},
|
| 864 |
+
{
|
| 865 |
+
"epoch": 0.9101251422070534,
|
| 866 |
+
"grad_norm": 35.27056121826172,
|
| 867 |
+
"learning_rate": 7.10447531142664e-06,
|
| 868 |
+
"loss": 0.6278,
|
| 869 |
+
"step": 550
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 0.9101251422070534,
|
| 873 |
+
"eval_accuracy": 0.6827906976744186,
|
| 874 |
+
"eval_loss": 0.6337892413139343,
|
| 875 |
+
"eval_runtime": 139.0169,
|
| 876 |
+
"eval_samples_per_second": 7.733,
|
| 877 |
+
"eval_steps_per_second": 7.733,
|
| 878 |
+
"step": 550
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.9183990071362085,
|
| 882 |
+
"grad_norm": 44.5506591796875,
|
| 883 |
+
"learning_rate": 7.0725657433052645e-06,
|
| 884 |
+
"loss": 0.6628,
|
| 885 |
+
"step": 555
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.9266728720653635,
|
| 889 |
+
"grad_norm": 164.92425537109375,
|
| 890 |
+
"learning_rate": 7.040462850579625e-06,
|
| 891 |
+
"loss": 0.5611,
|
| 892 |
+
"step": 560
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.9349467369945186,
|
| 896 |
+
"grad_norm": 64.9032211303711,
|
| 897 |
+
"learning_rate": 7.0081690457357975e-06,
|
| 898 |
+
"loss": 0.6663,
|
| 899 |
+
"step": 565
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.9432206019236736,
|
| 903 |
+
"grad_norm": 210.38734436035156,
|
| 904 |
+
"learning_rate": 6.975686755606624e-06,
|
| 905 |
+
"loss": 0.659,
|
| 906 |
+
"step": 570
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.9514944668528287,
|
| 910 |
+
"grad_norm": 60.48363494873047,
|
| 911 |
+
"learning_rate": 6.943018421189348e-06,
|
| 912 |
+
"loss": 0.597,
|
| 913 |
+
"step": 575
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.9597683317819836,
|
| 917 |
+
"grad_norm": 115.6025619506836,
|
| 918 |
+
"learning_rate": 6.910166497462174e-06,
|
| 919 |
+
"loss": 0.6677,
|
| 920 |
+
"step": 580
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.9680421967111387,
|
| 924 |
+
"grad_norm": 36.28033447265625,
|
| 925 |
+
"learning_rate": 6.877133453199773e-06,
|
| 926 |
+
"loss": 0.6969,
|
| 927 |
+
"step": 585
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.9763160616402937,
|
| 931 |
+
"grad_norm": 26.922504425048828,
|
| 932 |
+
"learning_rate": 6.843921770787765e-06,
|
| 933 |
+
"loss": 0.5802,
|
| 934 |
+
"step": 590
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.9845899265694488,
|
| 938 |
+
"grad_norm": 44.03938293457031,
|
| 939 |
+
"learning_rate": 6.810533946036172e-06,
|
| 940 |
+
"loss": 0.5862,
|
| 941 |
+
"step": 595
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.9928637914986038,
|
| 945 |
+
"grad_norm": 33.03104019165039,
|
| 946 |
+
"learning_rate": 6.7769724879918564e-06,
|
| 947 |
+
"loss": 0.6086,
|
| 948 |
+
"step": 600
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.9928637914986038,
|
| 952 |
+
"eval_accuracy": 0.6874418604651162,
|
| 953 |
+
"eval_loss": 0.6202675104141235,
|
| 954 |
+
"eval_runtime": 139.0223,
|
| 955 |
+
"eval_samples_per_second": 7.733,
|
| 956 |
+
"eval_steps_per_second": 7.733,
|
| 957 |
+
"step": 600
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 1.0011376564277588,
|
| 961 |
+
"grad_norm": 87.3049087524414,
|
| 962 |
+
"learning_rate": 6.743239918749973e-06,
|
| 963 |
+
"loss": 0.5948,
|
| 964 |
+
"step": 605
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"epoch": 1.009411521356914,
|
| 968 |
+
"grad_norm": 57.935916900634766,
|
| 969 |
+
"learning_rate": 6.709338773264435e-06,
|
| 970 |
+
"loss": 0.6225,
|
| 971 |
+
"step": 610
|
| 972 |
+
},
|
| 973 |
+
{
|
| 974 |
+
"epoch": 1.0176853862860689,
|
| 975 |
+
"grad_norm": 87.15520477294922,
|
| 976 |
+
"learning_rate": 6.675271599157415e-06,
|
| 977 |
+
"loss": 0.5865,
|
| 978 |
+
"step": 615
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 1.025959251215224,
|
| 982 |
+
"grad_norm": 35.6851806640625,
|
| 983 |
+
"learning_rate": 6.6410409565279035e-06,
|
| 984 |
+
"loss": 0.5406,
|
| 985 |
+
"step": 620
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 1.034233116144379,
|
| 989 |
+
"grad_norm": 110.91497802734375,
|
| 990 |
+
"learning_rate": 6.60664941775931e-06,
|
| 991 |
+
"loss": 0.6117,
|
| 992 |
+
"step": 625
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 1.042506981073534,
|
| 996 |
+
"grad_norm": 196.0850372314453,
|
| 997 |
+
"learning_rate": 6.572099567326157e-06,
|
| 998 |
+
"loss": 0.5984,
|
| 999 |
+
"step": 630
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 1.050780846002689,
|
| 1003 |
+
"grad_norm": 61.60392379760742,
|
| 1004 |
+
"learning_rate": 6.53739400159986e-06,
|
| 1005 |
+
"loss": 0.5954,
|
| 1006 |
+
"step": 635
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 1.059054710931844,
|
| 1010 |
+
"grad_norm": 81.43512725830078,
|
| 1011 |
+
"learning_rate": 6.502535328653616e-06,
|
| 1012 |
+
"loss": 0.6445,
|
| 1013 |
+
"step": 640
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 1.0673285758609992,
|
| 1017 |
+
"grad_norm": 84.67086791992188,
|
| 1018 |
+
"learning_rate": 6.467526168066408e-06,
|
| 1019 |
+
"loss": 0.5913,
|
| 1020 |
+
"step": 645
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 1.075602440790154,
|
| 1024 |
+
"grad_norm": 58.613800048828125,
|
| 1025 |
+
"learning_rate": 6.432369150726147e-06,
|
| 1026 |
+
"loss": 0.5708,
|
| 1027 |
+
"step": 650
|
| 1028 |
+
},
|
| 1029 |
+
{
|
| 1030 |
+
"epoch": 1.075602440790154,
|
| 1031 |
+
"eval_accuracy": 0.6911627906976744,
|
| 1032 |
+
"eval_loss": 0.6270654201507568,
|
| 1033 |
+
"eval_runtime": 135.6532,
|
| 1034 |
+
"eval_samples_per_second": 7.925,
|
| 1035 |
+
"eval_steps_per_second": 7.925,
|
| 1036 |
+
"step": 650
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 1.083876305719309,
|
| 1040 |
+
"grad_norm": 150.11831665039062,
|
| 1041 |
+
"learning_rate": 6.397066918631964e-06,
|
| 1042 |
+
"loss": 0.6169,
|
| 1043 |
+
"step": 655
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 1.0921501706484642,
|
| 1047 |
+
"grad_norm": 40.26316452026367,
|
| 1048 |
+
"learning_rate": 6.361622124695677e-06,
|
| 1049 |
+
"loss": 0.5921,
|
| 1050 |
+
"step": 660
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 1.1004240355776191,
|
| 1054 |
+
"grad_norm": 139.96380615234375,
|
| 1055 |
+
"learning_rate": 6.326037432542411e-06,
|
| 1056 |
+
"loss": 0.6532,
|
| 1057 |
+
"step": 665
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"epoch": 1.1086979005067743,
|
| 1061 |
+
"grad_norm": 75.26846313476562,
|
| 1062 |
+
"learning_rate": 6.290315516310446e-06,
|
| 1063 |
+
"loss": 0.671,
|
| 1064 |
+
"step": 670
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 1.1169717654359292,
|
| 1068 |
+
"grad_norm": 67.06864166259766,
|
| 1069 |
+
"learning_rate": 6.254459060450252e-06,
|
| 1070 |
+
"loss": 0.6279,
|
| 1071 |
+
"step": 675
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 1.1252456303650842,
|
| 1075 |
+
"grad_norm": 31.252126693725586,
|
| 1076 |
+
"learning_rate": 6.21847075952276e-06,
|
| 1077 |
+
"loss": 0.4873,
|
| 1078 |
+
"step": 680
|
| 1079 |
+
},
|
| 1080 |
+
{
|
| 1081 |
+
"epoch": 1.1335194952942393,
|
| 1082 |
+
"grad_norm": 26.07908058166504,
|
| 1083 |
+
"learning_rate": 6.1823533179968615e-06,
|
| 1084 |
+
"loss": 0.5434,
|
| 1085 |
+
"step": 685
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"epoch": 1.1417933602233943,
|
| 1089 |
+
"grad_norm": 188.46253967285156,
|
| 1090 |
+
"learning_rate": 6.146109450046187e-06,
|
| 1091 |
+
"loss": 0.6456,
|
| 1092 |
+
"step": 690
|
| 1093 |
+
},
|
| 1094 |
+
{
|
| 1095 |
+
"epoch": 1.1500672251525494,
|
| 1096 |
+
"grad_norm": 51.78955078125,
|
| 1097 |
+
"learning_rate": 6.1097418793451195e-06,
|
| 1098 |
+
"loss": 0.469,
|
| 1099 |
+
"step": 695
|
| 1100 |
+
},
|
| 1101 |
+
{
|
| 1102 |
+
"epoch": 1.1583410900817044,
|
| 1103 |
+
"grad_norm": 115.07262420654297,
|
| 1104 |
+
"learning_rate": 6.073253338864137e-06,
|
| 1105 |
+
"loss": 0.5732,
|
| 1106 |
+
"step": 700
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 1.1583410900817044,
|
| 1110 |
+
"eval_accuracy": 0.6865116279069767,
|
| 1111 |
+
"eval_loss": 0.6430281400680542,
|
| 1112 |
+
"eval_runtime": 135.7675,
|
| 1113 |
+
"eval_samples_per_second": 7.918,
|
| 1114 |
+
"eval_steps_per_second": 7.918,
|
| 1115 |
+
"step": 700
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 1.1666149550108595,
|
| 1119 |
+
"grad_norm": 137.6046905517578,
|
| 1120 |
+
"learning_rate": 6.036646570664412e-06,
|
| 1121 |
+
"loss": 0.5884,
|
| 1122 |
+
"step": 705
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 1.1748888199400145,
|
| 1126 |
+
"grad_norm": 89.0150146484375,
|
| 1127 |
+
"learning_rate": 5.999924325691765e-06,
|
| 1128 |
+
"loss": 0.583,
|
| 1129 |
+
"step": 710
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 1.1831626848691694,
|
| 1133 |
+
"grad_norm": 69.22703552246094,
|
| 1134 |
+
"learning_rate": 5.963089363569928e-06,
|
| 1135 |
+
"loss": 0.5935,
|
| 1136 |
+
"step": 715
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 1.1914365497983246,
|
| 1140 |
+
"grad_norm": 143.3881072998047,
|
| 1141 |
+
"learning_rate": 5.926144452393163e-06,
|
| 1142 |
+
"loss": 0.6591,
|
| 1143 |
+
"step": 720
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 1.1997104147274795,
|
| 1147 |
+
"grad_norm": 33.42167282104492,
|
| 1148 |
+
"learning_rate": 5.8890923685182464e-06,
|
| 1149 |
+
"loss": 0.5891,
|
| 1150 |
+
"step": 725
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 1.2079842796566347,
|
| 1154 |
+
"grad_norm": 79.74076843261719,
|
| 1155 |
+
"learning_rate": 5.851935896355828e-06,
|
| 1156 |
+
"loss": 0.6158,
|
| 1157 |
+
"step": 730
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 1.2162581445857896,
|
| 1161 |
+
"grad_norm": 68.50188446044922,
|
| 1162 |
+
"learning_rate": 5.814677828161186e-06,
|
| 1163 |
+
"loss": 0.5299,
|
| 1164 |
+
"step": 735
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 1.2245320095149448,
|
| 1168 |
+
"grad_norm": 110.22908782958984,
|
| 1169 |
+
"learning_rate": 5.7773209638243965e-06,
|
| 1170 |
+
"loss": 0.6114,
|
| 1171 |
+
"step": 740
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 1.2328058744440997,
|
| 1175 |
+
"grad_norm": 126.0649642944336,
|
| 1176 |
+
"learning_rate": 5.739868110659916e-06,
|
| 1177 |
+
"loss": 0.5822,
|
| 1178 |
+
"step": 745
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 1.2410797393732547,
|
| 1182 |
+
"grad_norm": 78.34774017333984,
|
| 1183 |
+
"learning_rate": 5.7023220831956335e-06,
|
| 1184 |
+
"loss": 0.5251,
|
| 1185 |
+
"step": 750
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 1.2410797393732547,
|
| 1189 |
+
"eval_accuracy": 0.6865116279069767,
|
| 1190 |
+
"eval_loss": 0.6465556025505066,
|
| 1191 |
+
"eval_runtime": 134.8395,
|
| 1192 |
+
"eval_samples_per_second": 7.972,
|
| 1193 |
+
"eval_steps_per_second": 7.972,
|
| 1194 |
+
"step": 750
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 1.2493536043024098,
|
| 1198 |
+
"grad_norm": 44.90232849121094,
|
| 1199 |
+
"learning_rate": 5.664685702961344e-06,
|
| 1200 |
+
"loss": 0.5814,
|
| 1201 |
+
"step": 755
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"epoch": 1.2576274692315648,
|
| 1205 |
+
"grad_norm": 61.099143981933594,
|
| 1206 |
+
"learning_rate": 5.626961798276727e-06,
|
| 1207 |
+
"loss": 0.6314,
|
| 1208 |
+
"step": 760
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 1.26590133416072,
|
| 1212 |
+
"grad_norm": 96.56782531738281,
|
| 1213 |
+
"learning_rate": 5.589153204038793e-06,
|
| 1214 |
+
"loss": 0.6495,
|
| 1215 |
+
"step": 765
|
| 1216 |
+
},
|
| 1217 |
+
{
|
| 1218 |
+
"epoch": 1.2741751990898749,
|
| 1219 |
+
"grad_norm": 96.17337799072266,
|
| 1220 |
+
"learning_rate": 5.551262761508857e-06,
|
| 1221 |
+
"loss": 0.5472,
|
| 1222 |
+
"step": 770
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"epoch": 1.2824490640190298,
|
| 1226 |
+
"grad_norm": 134.74688720703125,
|
| 1227 |
+
"learning_rate": 5.513293318099013e-06,
|
| 1228 |
+
"loss": 0.5901,
|
| 1229 |
+
"step": 775
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 1.290722928948185,
|
| 1233 |
+
"grad_norm": 58.68545913696289,
|
| 1234 |
+
"learning_rate": 5.475247727158154e-06,
|
| 1235 |
+
"loss": 0.625,
|
| 1236 |
+
"step": 780
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 1.29899679387734,
|
| 1240 |
+
"grad_norm": 44.06023406982422,
|
| 1241 |
+
"learning_rate": 5.437128847757554e-06,
|
| 1242 |
+
"loss": 0.5603,
|
| 1243 |
+
"step": 785
|
| 1244 |
+
},
|
| 1245 |
+
{
|
| 1246 |
+
"epoch": 1.307270658806495,
|
| 1247 |
+
"grad_norm": 79.8021011352539,
|
| 1248 |
+
"learning_rate": 5.398939544476004e-06,
|
| 1249 |
+
"loss": 0.5608,
|
| 1250 |
+
"step": 790
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 1.31554452373565,
|
| 1254 |
+
"grad_norm": 163.2118377685547,
|
| 1255 |
+
"learning_rate": 5.360682687184554e-06,
|
| 1256 |
+
"loss": 0.5516,
|
| 1257 |
+
"step": 795
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 1.323818388664805,
|
| 1261 |
+
"grad_norm": 133.09230041503906,
|
| 1262 |
+
"learning_rate": 5.3223611508308385e-06,
|
| 1263 |
+
"loss": 0.565,
|
| 1264 |
+
"step": 800
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 1.323818388664805,
|
| 1268 |
+
"eval_accuracy": 0.6902325581395349,
|
| 1269 |
+
"eval_loss": 0.6557974219322205,
|
| 1270 |
+
"eval_runtime": 136.9094,
|
| 1271 |
+
"eval_samples_per_second": 7.852,
|
| 1272 |
+
"eval_steps_per_second": 7.852,
|
| 1273 |
+
"step": 800
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 1.33209225359396,
|
| 1277 |
+
"grad_norm": 269.4532165527344,
|
| 1278 |
+
"learning_rate": 5.2839778152230266e-06,
|
| 1279 |
+
"loss": 0.6325,
|
| 1280 |
+
"step": 805
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 1.340366118523115,
|
| 1284 |
+
"grad_norm": 31.07293701171875,
|
| 1285 |
+
"learning_rate": 5.2455355648134174e-06,
|
| 1286 |
+
"loss": 0.6786,
|
| 1287 |
+
"step": 810
|
| 1288 |
+
},
|
| 1289 |
+
{
|
| 1290 |
+
"epoch": 1.3486399834522702,
|
| 1291 |
+
"grad_norm": 110.39642333984375,
|
| 1292 |
+
"learning_rate": 5.207037288481668e-06,
|
| 1293 |
+
"loss": 0.707,
|
| 1294 |
+
"step": 815
|
| 1295 |
+
},
|
| 1296 |
+
{
|
| 1297 |
+
"epoch": 1.3569138483814251,
|
| 1298 |
+
"grad_norm": 119.31515502929688,
|
| 1299 |
+
"learning_rate": 5.168485879317707e-06,
|
| 1300 |
+
"loss": 0.6234,
|
| 1301 |
+
"step": 820
|
| 1302 |
+
},
|
| 1303 |
+
{
|
| 1304 |
+
"epoch": 1.36518771331058,
|
| 1305 |
+
"grad_norm": 152.03526306152344,
|
| 1306 |
+
"learning_rate": 5.129884234404314e-06,
|
| 1307 |
+
"loss": 0.6802,
|
| 1308 |
+
"step": 825
|
| 1309 |
+
},
|
| 1310 |
+
{
|
| 1311 |
+
"epoch": 1.3734615782397352,
|
| 1312 |
+
"grad_norm": 85.00423431396484,
|
| 1313 |
+
"learning_rate": 5.091235254599416e-06,
|
| 1314 |
+
"loss": 0.5709,
|
| 1315 |
+
"step": 830
|
| 1316 |
+
},
|
| 1317 |
+
{
|
| 1318 |
+
"epoch": 1.3817354431688902,
|
| 1319 |
+
"grad_norm": 160.42672729492188,
|
| 1320 |
+
"learning_rate": 5.052541844318089e-06,
|
| 1321 |
+
"loss": 0.5747,
|
| 1322 |
+
"step": 835
|
| 1323 |
+
},
|
| 1324 |
+
{
|
| 1325 |
+
"epoch": 1.3900093080980453,
|
| 1326 |
+
"grad_norm": 57.966976165771484,
|
| 1327 |
+
"learning_rate": 5.013806911314294e-06,
|
| 1328 |
+
"loss": 0.5937,
|
| 1329 |
+
"step": 840
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 1.3982831730272003,
|
| 1333 |
+
"grad_norm": 39.06420135498047,
|
| 1334 |
+
"learning_rate": 4.975033366462361e-06,
|
| 1335 |
+
"loss": 0.6496,
|
| 1336 |
+
"step": 845
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 1.4065570379563552,
|
| 1340 |
+
"grad_norm": 124.18098449707031,
|
| 1341 |
+
"learning_rate": 4.936224123538254e-06,
|
| 1342 |
+
"loss": 0.6163,
|
| 1343 |
+
"step": 850
|
| 1344 |
+
},
|
| 1345 |
+
{
|
| 1346 |
+
"epoch": 1.4065570379563552,
|
| 1347 |
+
"eval_accuracy": 0.6809302325581396,
|
| 1348 |
+
"eval_loss": 0.6283326745033264,
|
| 1349 |
+
"eval_runtime": 135.7927,
|
| 1350 |
+
"eval_samples_per_second": 7.916,
|
| 1351 |
+
"eval_steps_per_second": 7.916,
|
| 1352 |
+
"step": 850
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 1.4148309028855104,
|
| 1356 |
+
"grad_norm": 109.61412048339844,
|
| 1357 |
+
"learning_rate": 4.897382099000587e-06,
|
| 1358 |
+
"loss": 0.5904,
|
| 1359 |
+
"step": 855
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 1.4231047678146655,
|
| 1363 |
+
"grad_norm": 97.67943572998047,
|
| 1364 |
+
"learning_rate": 4.858510211771469e-06,
|
| 1365 |
+
"loss": 0.6084,
|
| 1366 |
+
"step": 860
|
| 1367 |
+
},
|
| 1368 |
+
{
|
| 1369 |
+
"epoch": 1.4313786327438205,
|
| 1370 |
+
"grad_norm": 32.73789978027344,
|
| 1371 |
+
"learning_rate": 4.819611383017145e-06,
|
| 1372 |
+
"loss": 0.5371,
|
| 1373 |
+
"step": 865
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 1.4396524976729754,
|
| 1377 |
+
"grad_norm": 37.779815673828125,
|
| 1378 |
+
"learning_rate": 4.780688535928477e-06,
|
| 1379 |
+
"loss": 0.5433,
|
| 1380 |
+
"step": 870
|
| 1381 |
+
},
|
| 1382 |
+
{
|
| 1383 |
+
"epoch": 1.4479263626021306,
|
| 1384 |
+
"grad_norm": 65.8934326171875,
|
| 1385 |
+
"learning_rate": 4.741744595501275e-06,
|
| 1386 |
+
"loss": 0.5889,
|
| 1387 |
+
"step": 875
|
| 1388 |
+
},
|
| 1389 |
+
{
|
| 1390 |
+
"epoch": 1.4562002275312855,
|
| 1391 |
+
"grad_norm": 94.01367950439453,
|
| 1392 |
+
"learning_rate": 4.702782488316478e-06,
|
| 1393 |
+
"loss": 0.5833,
|
| 1394 |
+
"step": 880
|
| 1395 |
+
},
|
| 1396 |
+
{
|
| 1397 |
+
"epoch": 1.4644740924604407,
|
| 1398 |
+
"grad_norm": 100.82835388183594,
|
| 1399 |
+
"learning_rate": 4.66380514232023e-06,
|
| 1400 |
+
"loss": 0.6123,
|
| 1401 |
+
"step": 885
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"epoch": 1.4727479573895956,
|
| 1405 |
+
"grad_norm": 118.06622314453125,
|
| 1406 |
+
"learning_rate": 4.6248154866038504e-06,
|
| 1407 |
+
"loss": 0.5869,
|
| 1408 |
+
"step": 890
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 1.4810218223187506,
|
| 1412 |
+
"grad_norm": 80.13727569580078,
|
| 1413 |
+
"learning_rate": 4.585816451183721e-06,
|
| 1414 |
+
"loss": 0.5605,
|
| 1415 |
+
"step": 895
|
| 1416 |
+
},
|
| 1417 |
+
{
|
| 1418 |
+
"epoch": 1.4892956872479057,
|
| 1419 |
+
"grad_norm": 64.97086334228516,
|
| 1420 |
+
"learning_rate": 4.546810966781089e-06,
|
| 1421 |
+
"loss": 0.5409,
|
| 1422 |
+
"step": 900
|
| 1423 |
+
},
|
| 1424 |
+
{
|
| 1425 |
+
"epoch": 1.4892956872479057,
|
| 1426 |
+
"eval_accuracy": 0.6893023255813954,
|
| 1427 |
+
"eval_loss": 0.6331726908683777,
|
| 1428 |
+
"eval_runtime": 136.5834,
|
| 1429 |
+
"eval_samples_per_second": 7.871,
|
| 1430 |
+
"eval_steps_per_second": 7.871,
|
| 1431 |
+
"step": 900
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 1.4975695521770607,
|
| 1435 |
+
"grad_norm": 70.27656555175781,
|
| 1436 |
+
"learning_rate": 4.507801964601837e-06,
|
| 1437 |
+
"loss": 0.543,
|
| 1438 |
+
"step": 905
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 1.5058434171062158,
|
| 1442 |
+
"grad_norm": 44.15786361694336,
|
| 1443 |
+
"learning_rate": 4.468792376116199e-06,
|
| 1444 |
+
"loss": 0.6316,
|
| 1445 |
+
"step": 910
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 1.5141172820353708,
|
| 1449 |
+
"grad_norm": 26.016767501831055,
|
| 1450 |
+
"learning_rate": 4.429785132838475e-06,
|
| 1451 |
+
"loss": 0.5746,
|
| 1452 |
+
"step": 915
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 1.5223911469645257,
|
| 1456 |
+
"grad_norm": 121.1869888305664,
|
| 1457 |
+
"learning_rate": 4.390783166106721e-06,
|
| 1458 |
+
"loss": 0.7063,
|
| 1459 |
+
"step": 920
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 1.5306650118936809,
|
| 1463 |
+
"grad_norm": 45.07545852661133,
|
| 1464 |
+
"learning_rate": 4.3517894068624704e-06,
|
| 1465 |
+
"loss": 0.5817,
|
| 1466 |
+
"step": 925
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 1.538938876822836,
|
| 1470 |
+
"grad_norm": 87.12654876708984,
|
| 1471 |
+
"learning_rate": 4.312806785430477e-06,
|
| 1472 |
+
"loss": 0.6364,
|
| 1473 |
+
"step": 930
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 1.547212741751991,
|
| 1477 |
+
"grad_norm": 66.20915222167969,
|
| 1478 |
+
"learning_rate": 4.273838231298501e-06,
|
| 1479 |
+
"loss": 0.5762,
|
| 1480 |
+
"step": 935
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 1.555486606681146,
|
| 1484 |
+
"grad_norm": 91.70149993896484,
|
| 1485 |
+
"learning_rate": 4.2348866728971695e-06,
|
| 1486 |
+
"loss": 0.6013,
|
| 1487 |
+
"step": 940
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 1.5637604716103009,
|
| 1491 |
+
"grad_norm": 61.58055877685547,
|
| 1492 |
+
"learning_rate": 4.195955037379899e-06,
|
| 1493 |
+
"loss": 0.6175,
|
| 1494 |
+
"step": 945
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 1.572034336539456,
|
| 1498 |
+
"grad_norm": 208.4636993408203,
|
| 1499 |
+
"learning_rate": 4.157046250402935e-06,
|
| 1500 |
+
"loss": 0.6015,
|
| 1501 |
+
"step": 950
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 1.572034336539456,
|
| 1505 |
+
"eval_accuracy": 0.6902325581395349,
|
| 1506 |
+
"eval_loss": 0.621203601360321,
|
| 1507 |
+
"eval_runtime": 135.2354,
|
| 1508 |
+
"eval_samples_per_second": 7.949,
|
| 1509 |
+
"eval_steps_per_second": 7.949,
|
| 1510 |
+
"step": 950
|
| 1511 |
+
},
|
| 1512 |
+
{
|
| 1513 |
+
"epoch": 1.5803082014686112,
|
| 1514 |
+
"grad_norm": 26.804574966430664,
|
| 1515 |
+
"learning_rate": 4.118163235905488e-06,
|
| 1516 |
+
"loss": 0.5719,
|
| 1517 |
+
"step": 955
|
| 1518 |
+
},
|
| 1519 |
+
{
|
| 1520 |
+
"epoch": 1.5885820663977661,
|
| 1521 |
+
"grad_norm": 128.2564239501953,
|
| 1522 |
+
"learning_rate": 4.079308915889999e-06,
|
| 1523 |
+
"loss": 0.5666,
|
| 1524 |
+
"step": 960
|
| 1525 |
+
},
|
| 1526 |
+
{
|
| 1527 |
+
"epoch": 1.596855931326921,
|
| 1528 |
+
"grad_norm": 69.91098022460938,
|
| 1529 |
+
"learning_rate": 4.040486210202567e-06,
|
| 1530 |
+
"loss": 0.5785,
|
| 1531 |
+
"step": 965
|
| 1532 |
+
},
|
| 1533 |
+
{
|
| 1534 |
+
"epoch": 1.605129796256076,
|
| 1535 |
+
"grad_norm": 134.2613067626953,
|
| 1536 |
+
"learning_rate": 4.001698036313514e-06,
|
| 1537 |
+
"loss": 0.6613,
|
| 1538 |
+
"step": 970
|
| 1539 |
+
},
|
| 1540 |
+
{
|
| 1541 |
+
"epoch": 1.6134036611852312,
|
| 1542 |
+
"grad_norm": 104.1937255859375,
|
| 1543 |
+
"learning_rate": 3.96294730909815e-06,
|
| 1544 |
+
"loss": 0.5157,
|
| 1545 |
+
"step": 975
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 1.6216775261143863,
|
| 1549 |
+
"grad_norm": 75.18810272216797,
|
| 1550 |
+
"learning_rate": 3.924236940617723e-06,
|
| 1551 |
+
"loss": 0.6356,
|
| 1552 |
+
"step": 980
|
| 1553 |
+
},
|
| 1554 |
+
{
|
| 1555 |
+
"epoch": 1.6299513910435413,
|
| 1556 |
+
"grad_norm": 35.88945007324219,
|
| 1557 |
+
"learning_rate": 3.885569839900576e-06,
|
| 1558 |
+
"loss": 0.6173,
|
| 1559 |
+
"step": 985
|
| 1560 |
+
},
|
| 1561 |
+
{
|
| 1562 |
+
"epoch": 1.6382252559726962,
|
| 1563 |
+
"grad_norm": 29.352340698242188,
|
| 1564 |
+
"learning_rate": 3.846948912723542e-06,
|
| 1565 |
+
"loss": 0.6053,
|
| 1566 |
+
"step": 990
|
| 1567 |
+
},
|
| 1568 |
+
{
|
| 1569 |
+
"epoch": 1.6464991209018511,
|
| 1570 |
+
"grad_norm": 132.6302490234375,
|
| 1571 |
+
"learning_rate": 3.808377061393587e-06,
|
| 1572 |
+
"loss": 0.6604,
|
| 1573 |
+
"step": 995
|
| 1574 |
+
},
|
| 1575 |
+
{
|
| 1576 |
+
"epoch": 1.6547729858310063,
|
| 1577 |
+
"grad_norm": 158.6077423095703,
|
| 1578 |
+
"learning_rate": 3.7698571845296876e-06,
|
| 1579 |
+
"loss": 0.5921,
|
| 1580 |
+
"step": 1000
|
| 1581 |
+
},
|
| 1582 |
+
{
|
| 1583 |
+
"epoch": 1.6547729858310063,
|
| 1584 |
+
"eval_accuracy": 0.6893023255813954,
|
| 1585 |
+
"eval_loss": 0.6269645690917969,
|
| 1586 |
+
"eval_runtime": 132.1367,
|
| 1587 |
+
"eval_samples_per_second": 8.136,
|
| 1588 |
+
"eval_steps_per_second": 8.136,
|
| 1589 |
+
"step": 1000
|
| 1590 |
+
},
|
| 1591 |
+
{
|
| 1592 |
+
"epoch": 1.6630468507601615,
|
| 1593 |
+
"grad_norm": 64.79986572265625,
|
| 1594 |
+
"learning_rate": 3.731392176845023e-06,
|
| 1595 |
+
"loss": 0.5632,
|
| 1596 |
+
"step": 1005
|
| 1597 |
+
},
|
| 1598 |
+
{
|
| 1599 |
+
"epoch": 1.6713207156893164,
|
| 1600 |
+
"grad_norm": 69.55998229980469,
|
| 1601 |
+
"learning_rate": 3.692984928929426e-06,
|
| 1602 |
+
"loss": 0.5757,
|
| 1603 |
+
"step": 1010
|
| 1604 |
+
},
|
| 1605 |
+
{
|
| 1606 |
+
"epoch": 1.6795945806184713,
|
| 1607 |
+
"grad_norm": 58.388427734375,
|
| 1608 |
+
"learning_rate": 3.6546383270321673e-06,
|
| 1609 |
+
"loss": 0.5728,
|
| 1610 |
+
"step": 1015
|
| 1611 |
+
},
|
| 1612 |
+
{
|
| 1613 |
+
"epoch": 1.6878684455476263,
|
| 1614 |
+
"grad_norm": 209.67808532714844,
|
| 1615 |
+
"learning_rate": 3.6163552528450616e-06,
|
| 1616 |
+
"loss": 0.6385,
|
| 1617 |
+
"step": 1020
|
| 1618 |
+
},
|
| 1619 |
+
{
|
| 1620 |
+
"epoch": 1.6961423104767814,
|
| 1621 |
+
"grad_norm": 41.65730667114258,
|
| 1622 |
+
"learning_rate": 3.5781385832859032e-06,
|
| 1623 |
+
"loss": 0.5542,
|
| 1624 |
+
"step": 1025
|
| 1625 |
+
},
|
| 1626 |
+
{
|
| 1627 |
+
"epoch": 1.7044161754059366,
|
| 1628 |
+
"grad_norm": 40.03094482421875,
|
| 1629 |
+
"learning_rate": 3.539991190282274e-06,
|
| 1630 |
+
"loss": 0.5365,
|
| 1631 |
+
"step": 1030
|
| 1632 |
+
},
|
| 1633 |
+
{
|
| 1634 |
+
"epoch": 1.7126900403350915,
|
| 1635 |
+
"grad_norm": 243.15150451660156,
|
| 1636 |
+
"learning_rate": 3.501915940555728e-06,
|
| 1637 |
+
"loss": 0.7109,
|
| 1638 |
+
"step": 1035
|
| 1639 |
+
},
|
| 1640 |
+
{
|
| 1641 |
+
"epoch": 1.7209639052642465,
|
| 1642 |
+
"grad_norm": 42.63383865356445,
|
| 1643 |
+
"learning_rate": 3.4639156954063487e-06,
|
| 1644 |
+
"loss": 0.5262,
|
| 1645 |
+
"step": 1040
|
| 1646 |
+
},
|
| 1647 |
+
{
|
| 1648 |
+
"epoch": 1.7292377701934016,
|
| 1649 |
+
"grad_norm": 110.11820983886719,
|
| 1650 |
+
"learning_rate": 3.4259933104977394e-06,
|
| 1651 |
+
"loss": 0.581,
|
| 1652 |
+
"step": 1045
|
| 1653 |
+
},
|
| 1654 |
+
{
|
| 1655 |
+
"epoch": 1.7375116351225566,
|
| 1656 |
+
"grad_norm": 140.5242919921875,
|
| 1657 |
+
"learning_rate": 3.3881516356424178e-06,
|
| 1658 |
+
"loss": 0.5132,
|
| 1659 |
+
"step": 1050
|
| 1660 |
+
},
|
| 1661 |
+
{
|
| 1662 |
+
"epoch": 1.7375116351225566,
|
| 1663 |
+
"eval_accuracy": 0.6911627906976744,
|
| 1664 |
+
"eval_loss": 0.6242228150367737,
|
| 1665 |
+
"eval_runtime": 135.3409,
|
| 1666 |
+
"eval_samples_per_second": 7.943,
|
| 1667 |
+
"eval_steps_per_second": 7.943,
|
| 1668 |
+
"step": 1050
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 1.7457855000517117,
|
| 1672 |
+
"grad_norm": 117.73894500732422,
|
| 1673 |
+
"learning_rate": 3.3503935145876525e-06,
|
| 1674 |
+
"loss": 0.6865,
|
| 1675 |
+
"step": 1055
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 1.7540593649808667,
|
| 1679 |
+
"grad_norm": 64.21101379394531,
|
| 1680 |
+
"learning_rate": 3.312721784801774e-06,
|
| 1681 |
+
"loss": 0.5874,
|
| 1682 |
+
"step": 1060
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 1.7623332299100216,
|
| 1686 |
+
"grad_norm": 74.21529388427734,
|
| 1687 |
+
"learning_rate": 3.275139277260925e-06,
|
| 1688 |
+
"loss": 0.6284,
|
| 1689 |
+
"step": 1065
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 1.7706070948391768,
|
| 1693 |
+
"grad_norm": 63.436607360839844,
|
| 1694 |
+
"learning_rate": 3.2376488162363275e-06,
|
| 1695 |
+
"loss": 0.5562,
|
| 1696 |
+
"step": 1070
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 1.778880959768332,
|
| 1700 |
+
"grad_norm": 57.009483337402344,
|
| 1701 |
+
"learning_rate": 3.2002532190820422e-06,
|
| 1702 |
+
"loss": 0.5425,
|
| 1703 |
+
"step": 1075
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 1.7871548246974869,
|
| 1707 |
+
"grad_norm": 45.6854362487793,
|
| 1708 |
+
"learning_rate": 3.162955296023247e-06,
|
| 1709 |
+
"loss": 0.8092,
|
| 1710 |
+
"step": 1080
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 1.7954286896266418,
|
| 1714 |
+
"grad_norm": 68.54329681396484,
|
| 1715 |
+
"learning_rate": 3.125757849945051e-06,
|
| 1716 |
+
"loss": 0.6519,
|
| 1717 |
+
"step": 1085
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 1.8037025545557968,
|
| 1721 |
+
"grad_norm": 62.51036834716797,
|
| 1722 |
+
"learning_rate": 3.088663676181864e-06,
|
| 1723 |
+
"loss": 0.5576,
|
| 1724 |
+
"step": 1090
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 1.811976419484952,
|
| 1728 |
+
"grad_norm": 108.84650421142578,
|
| 1729 |
+
"learning_rate": 3.0516755623073273e-06,
|
| 1730 |
+
"loss": 0.5965,
|
| 1731 |
+
"step": 1095
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 1.820250284414107,
|
| 1735 |
+
"grad_norm": 39.56748580932617,
|
| 1736 |
+
"learning_rate": 3.0147962879248402e-06,
|
| 1737 |
+
"loss": 0.5851,
|
| 1738 |
+
"step": 1100
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 1.820250284414107,
|
| 1742 |
+
"eval_accuracy": 0.6874418604651162,
|
| 1743 |
+
"eval_loss": 0.6186655163764954,
|
| 1744 |
+
"eval_runtime": 132.3,
|
| 1745 |
+
"eval_samples_per_second": 8.125,
|
| 1746 |
+
"eval_steps_per_second": 8.125,
|
| 1747 |
+
"step": 1100
|
| 1748 |
+
},
|
| 1749 |
+
{
|
| 1750 |
+
"epoch": 1.828524149343262,
|
| 1751 |
+
"grad_norm": 115.91883087158203,
|
| 1752 |
+
"learning_rate": 2.978028624458668e-06,
|
| 1753 |
+
"loss": 0.5346,
|
| 1754 |
+
"step": 1105
|
| 1755 |
+
},
|
| 1756 |
+
{
|
| 1757 |
+
"epoch": 1.836798014272417,
|
| 1758 |
+
"grad_norm": 58.3183708190918,
|
| 1759 |
+
"learning_rate": 2.941375334945674e-06,
|
| 1760 |
+
"loss": 0.5022,
|
| 1761 |
+
"step": 1110
|
| 1762 |
+
},
|
| 1763 |
+
{
|
| 1764 |
+
"epoch": 1.845071879201572,
|
| 1765 |
+
"grad_norm": 80.68195343017578,
|
| 1766 |
+
"learning_rate": 2.9048391738276908e-06,
|
| 1767 |
+
"loss": 0.6279,
|
| 1768 |
+
"step": 1115
|
| 1769 |
+
},
|
| 1770 |
+
{
|
| 1771 |
+
"epoch": 1.853345744130727,
|
| 1772 |
+
"grad_norm": 71.54075622558594,
|
| 1773 |
+
"learning_rate": 2.8684228867445133e-06,
|
| 1774 |
+
"loss": 0.5217,
|
| 1775 |
+
"step": 1120
|
| 1776 |
+
},
|
| 1777 |
+
{
|
| 1778 |
+
"epoch": 1.8616196090598822,
|
| 1779 |
+
"grad_norm": 45.025848388671875,
|
| 1780 |
+
"learning_rate": 2.8321292103275813e-06,
|
| 1781 |
+
"loss": 0.5751,
|
| 1782 |
+
"step": 1125
|
| 1783 |
+
},
|
| 1784 |
+
{
|
| 1785 |
+
"epoch": 1.8698934739890372,
|
| 1786 |
+
"grad_norm": 119.63114929199219,
|
| 1787 |
+
"learning_rate": 2.7959608719943193e-06,
|
| 1788 |
+
"loss": 0.5505,
|
| 1789 |
+
"step": 1130
|
| 1790 |
+
},
|
| 1791 |
+
{
|
| 1792 |
+
"epoch": 1.878167338918192,
|
| 1793 |
+
"grad_norm": 29.393390655517578,
|
| 1794 |
+
"learning_rate": 2.7599205897431707e-06,
|
| 1795 |
+
"loss": 0.5426,
|
| 1796 |
+
"step": 1135
|
| 1797 |
+
},
|
| 1798 |
+
{
|
| 1799 |
+
"epoch": 1.886441203847347,
|
| 1800 |
+
"grad_norm": 86.28485870361328,
|
| 1801 |
+
"learning_rate": 2.7240110719493568e-06,
|
| 1802 |
+
"loss": 0.5846,
|
| 1803 |
+
"step": 1140
|
| 1804 |
+
},
|
| 1805 |
+
{
|
| 1806 |
+
"epoch": 1.8947150687765022,
|
| 1807 |
+
"grad_norm": 102.7513656616211,
|
| 1808 |
+
"learning_rate": 2.688235017161331e-06,
|
| 1809 |
+
"loss": 0.6677,
|
| 1810 |
+
"step": 1145
|
| 1811 |
+
},
|
| 1812 |
+
{
|
| 1813 |
+
"epoch": 1.9029889337056574,
|
| 1814 |
+
"grad_norm": 45.93974685668945,
|
| 1815 |
+
"learning_rate": 2.6525951138979987e-06,
|
| 1816 |
+
"loss": 0.5869,
|
| 1817 |
+
"step": 1150
|
| 1818 |
+
},
|
| 1819 |
+
{
|
| 1820 |
+
"epoch": 1.9029889337056574,
|
| 1821 |
+
"eval_accuracy": 0.6827906976744186,
|
| 1822 |
+
"eval_loss": 0.6229726076126099,
|
| 1823 |
+
"eval_runtime": 142.7285,
|
| 1824 |
+
"eval_samples_per_second": 7.532,
|
| 1825 |
+
"eval_steps_per_second": 7.532,
|
| 1826 |
+
"step": 1150
|
| 1827 |
+
},
|
| 1828 |
+
{
|
| 1829 |
+
"epoch": 1.9112627986348123,
|
| 1830 |
+
"grad_norm": 39.167236328125,
|
| 1831 |
+
"learning_rate": 2.617094040446676e-06,
|
| 1832 |
+
"loss": 0.5982,
|
| 1833 |
+
"step": 1155
|
| 1834 |
+
},
|
| 1835 |
+
{
|
| 1836 |
+
"epoch": 1.9195366635639672,
|
| 1837 |
+
"grad_norm": 161.03598022460938,
|
| 1838 |
+
"learning_rate": 2.5817344646618134e-06,
|
| 1839 |
+
"loss": 0.6664,
|
| 1840 |
+
"step": 1160
|
| 1841 |
+
},
|
| 1842 |
+
{
|
| 1843 |
+
"epoch": 1.9278105284931224,
|
| 1844 |
+
"grad_norm": 94.9665298461914,
|
| 1845 |
+
"learning_rate": 2.5465190437645224e-06,
|
| 1846 |
+
"loss": 0.5211,
|
| 1847 |
+
"step": 1165
|
| 1848 |
+
},
|
| 1849 |
+
{
|
| 1850 |
+
"epoch": 1.9360843934222773,
|
| 1851 |
+
"grad_norm": 36.561275482177734,
|
| 1852 |
+
"learning_rate": 2.511450424142878e-06,
|
| 1853 |
+
"loss": 0.6397,
|
| 1854 |
+
"step": 1170
|
| 1855 |
+
},
|
| 1856 |
+
{
|
| 1857 |
+
"epoch": 1.9443582583514325,
|
| 1858 |
+
"grad_norm": 144.52769470214844,
|
| 1859 |
+
"learning_rate": 2.4765312411530504e-06,
|
| 1860 |
+
"loss": 0.6234,
|
| 1861 |
+
"step": 1175
|
| 1862 |
+
},
|
| 1863 |
+
{
|
| 1864 |
+
"epoch": 1.9526321232805874,
|
| 1865 |
+
"grad_norm": 70.71371459960938,
|
| 1866 |
+
"learning_rate": 2.4417641189212683e-06,
|
| 1867 |
+
"loss": 0.6121,
|
| 1868 |
+
"step": 1180
|
| 1869 |
+
},
|
| 1870 |
+
{
|
| 1871 |
+
"epoch": 1.9609059882097424,
|
| 1872 |
+
"grad_norm": 150.60153198242188,
|
| 1873 |
+
"learning_rate": 2.4071516701466072e-06,
|
| 1874 |
+
"loss": 0.5721,
|
| 1875 |
+
"step": 1185
|
| 1876 |
+
},
|
| 1877 |
+
{
|
| 1878 |
+
"epoch": 1.9691798531388975,
|
| 1879 |
+
"grad_norm": 65.08445739746094,
|
| 1880 |
+
"learning_rate": 2.37269649590466e-06,
|
| 1881 |
+
"loss": 0.574,
|
| 1882 |
+
"step": 1190
|
| 1883 |
+
},
|
| 1884 |
+
{
|
| 1885 |
+
"epoch": 1.9774537180680527,
|
| 1886 |
+
"grad_norm": 85.44747161865234,
|
| 1887 |
+
"learning_rate": 2.3384011854520643e-06,
|
| 1888 |
+
"loss": 0.55,
|
| 1889 |
+
"step": 1195
|
| 1890 |
+
},
|
| 1891 |
+
{
|
| 1892 |
+
"epoch": 1.9857275829972076,
|
| 1893 |
+
"grad_norm": 95.13154602050781,
|
| 1894 |
+
"learning_rate": 2.304268316031922e-06,
|
| 1895 |
+
"loss": 0.558,
|
| 1896 |
+
"step": 1200
|
| 1897 |
+
},
|
| 1898 |
+
{
|
| 1899 |
+
"epoch": 1.9857275829972076,
|
| 1900 |
+
"eval_accuracy": 0.693953488372093,
|
| 1901 |
+
"eval_loss": 0.6230591535568237,
|
| 1902 |
+
"eval_runtime": 139.4556,
|
| 1903 |
+
"eval_samples_per_second": 7.709,
|
| 1904 |
+
"eval_steps_per_second": 7.709,
|
| 1905 |
+
"step": 1200
|
| 1906 |
+
},
|
| 1907 |
+
{
|
| 1908 |
+
"epoch": 1.9940014479263626,
|
| 1909 |
+
"grad_norm": 72.64482879638672,
|
| 1910 |
+
"learning_rate": 2.2703004526801315e-06,
|
| 1911 |
+
"loss": 0.6093,
|
| 1912 |
+
"step": 1205
|
| 1913 |
+
},
|
| 1914 |
+
{
|
| 1915 |
+
"epoch": 2.0022753128555175,
|
| 1916 |
+
"grad_norm": 91.53067016601562,
|
| 1917 |
+
"learning_rate": 2.236500148032616e-06,
|
| 1918 |
+
"loss": 0.5926,
|
| 1919 |
+
"step": 1210
|
| 1920 |
+
},
|
| 1921 |
+
{
|
| 1922 |
+
"epoch": 2.0105491777846725,
|
| 1923 |
+
"grad_norm": 101.95372772216797,
|
| 1924 |
+
"learning_rate": 2.2028699421335074e-06,
|
| 1925 |
+
"loss": 0.5415,
|
| 1926 |
+
"step": 1215
|
| 1927 |
+
},
|
| 1928 |
+
{
|
| 1929 |
+
"epoch": 2.018823042713828,
|
| 1930 |
+
"grad_norm": 103.2565689086914,
|
| 1931 |
+
"learning_rate": 2.1694123622442645e-06,
|
| 1932 |
+
"loss": 0.5096,
|
| 1933 |
+
"step": 1220
|
| 1934 |
+
},
|
| 1935 |
+
{
|
| 1936 |
+
"epoch": 2.027096907642983,
|
| 1937 |
+
"grad_norm": 32.46419906616211,
|
| 1938 |
+
"learning_rate": 2.1361299226537465e-06,
|
| 1939 |
+
"loss": 0.4894,
|
| 1940 |
+
"step": 1225
|
| 1941 |
+
},
|
| 1942 |
+
{
|
| 1943 |
+
"epoch": 2.0353707725721377,
|
| 1944 |
+
"grad_norm": 59.69719696044922,
|
| 1945 |
+
"learning_rate": 2.1030251244892713e-06,
|
| 1946 |
+
"loss": 0.5322,
|
| 1947 |
+
"step": 1230
|
| 1948 |
+
},
|
| 1949 |
+
{
|
| 1950 |
+
"epoch": 2.0436446375012927,
|
| 1951 |
+
"grad_norm": 44.855892181396484,
|
| 1952 |
+
"learning_rate": 2.0701004555286637e-06,
|
| 1953 |
+
"loss": 0.5304,
|
| 1954 |
+
"step": 1235
|
| 1955 |
+
},
|
| 1956 |
+
{
|
| 1957 |
+
"epoch": 2.051918502430448,
|
| 1958 |
+
"grad_norm": 124.84123992919922,
|
| 1959 |
+
"learning_rate": 2.0373583900132974e-06,
|
| 1960 |
+
"loss": 0.5007,
|
| 1961 |
+
"step": 1240
|
| 1962 |
+
},
|
| 1963 |
+
{
|
| 1964 |
+
"epoch": 2.060192367359603,
|
| 1965 |
+
"grad_norm": 178.7327117919922,
|
| 1966 |
+
"learning_rate": 2.0048013884621617e-06,
|
| 1967 |
+
"loss": 0.5525,
|
| 1968 |
+
"step": 1245
|
| 1969 |
+
},
|
| 1970 |
+
{
|
| 1971 |
+
"epoch": 2.068466232288758,
|
| 1972 |
+
"grad_norm": 89.66055297851562,
|
| 1973 |
+
"learning_rate": 1.9724318974869516e-06,
|
| 1974 |
+
"loss": 0.5181,
|
| 1975 |
+
"step": 1250
|
| 1976 |
+
},
|
| 1977 |
+
{
|
| 1978 |
+
"epoch": 2.068466232288758,
|
| 1979 |
+
"eval_accuracy": 0.6920930232558139,
|
| 1980 |
+
"eval_loss": 0.6200288534164429,
|
| 1981 |
+
"eval_runtime": 139.1734,
|
| 1982 |
+
"eval_samples_per_second": 7.724,
|
| 1983 |
+
"eval_steps_per_second": 7.724,
|
| 1984 |
+
"step": 1250
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 2.076740097217913,
|
| 1988 |
+
"grad_norm": 144.76039123535156,
|
| 1989 |
+
"learning_rate": 1.940252349608215e-06,
|
| 1990 |
+
"loss": 0.5971,
|
| 1991 |
+
"step": 1255
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 2.085013962147068,
|
| 1995 |
+
"grad_norm": 78.81449890136719,
|
| 1996 |
+
"learning_rate": 1.908265163072554e-06,
|
| 1997 |
+
"loss": 0.6142,
|
| 1998 |
+
"step": 1260
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 2.093287827076223,
|
| 2002 |
+
"grad_norm": 161.04014587402344,
|
| 2003 |
+
"learning_rate": 1.8764727416708883e-06,
|
| 2004 |
+
"loss": 0.5488,
|
| 2005 |
+
"step": 1265
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 2.101561692005378,
|
| 2009 |
+
"grad_norm": 175.5146942138672,
|
| 2010 |
+
"learning_rate": 1.8448774745578186e-06,
|
| 2011 |
+
"loss": 0.6606,
|
| 2012 |
+
"step": 1270
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 2.109835556934533,
|
| 2016 |
+
"grad_norm": 62.92079544067383,
|
| 2017 |
+
"learning_rate": 1.81348173607209e-06,
|
| 2018 |
+
"loss": 0.5204,
|
| 2019 |
+
"step": 1275
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 2.118109421863688,
|
| 2023 |
+
"grad_norm": 150.65797424316406,
|
| 2024 |
+
"learning_rate": 1.7822878855581543e-06,
|
| 2025 |
+
"loss": 0.4931,
|
| 2026 |
+
"step": 1280
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 2.126383286792843,
|
| 2030 |
+
"grad_norm": 64.05603790283203,
|
| 2031 |
+
"learning_rate": 1.7512982671888757e-06,
|
| 2032 |
+
"loss": 0.5615,
|
| 2033 |
+
"step": 1285
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 2.1346571517219983,
|
| 2037 |
+
"grad_norm": 42.53824234008789,
|
| 2038 |
+
"learning_rate": 1.7205152097893694e-06,
|
| 2039 |
+
"loss": 0.5017,
|
| 2040 |
+
"step": 1290
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 2.1429310166511533,
|
| 2044 |
+
"grad_norm": 67.79627990722656,
|
| 2045 |
+
"learning_rate": 1.689941026661986e-06,
|
| 2046 |
+
"loss": 0.5048,
|
| 2047 |
+
"step": 1295
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 2.151204881580308,
|
| 2051 |
+
"grad_norm": 45.754302978515625,
|
| 2052 |
+
"learning_rate": 1.6595780154124827e-06,
|
| 2053 |
+
"loss": 0.5542,
|
| 2054 |
+
"step": 1300
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 2.151204881580308,
|
| 2058 |
+
"eval_accuracy": 0.6874418604651162,
|
| 2059 |
+
"eval_loss": 0.625083327293396,
|
| 2060 |
+
"eval_runtime": 139.2113,
|
| 2061 |
+
"eval_samples_per_second": 7.722,
|
| 2062 |
+
"eval_steps_per_second": 7.722,
|
| 2063 |
+
"step": 1300
|
| 2064 |
+
},
|
| 2065 |
+
{
|
| 2066 |
+
"epoch": 2.159478746509463,
|
| 2067 |
+
"grad_norm": 53.39175033569336,
|
| 2068 |
+
"learning_rate": 1.6294284577773493e-06,
|
| 2069 |
+
"loss": 0.5467,
|
| 2070 |
+
"step": 1305
|
| 2071 |
+
},
|
| 2072 |
+
{
|
| 2073 |
+
"epoch": 2.167752611438618,
|
| 2074 |
+
"grad_norm": 135.67037963867188,
|
| 2075 |
+
"learning_rate": 1.5994946194523452e-06,
|
| 2076 |
+
"loss": 0.5632,
|
| 2077 |
+
"step": 1310
|
| 2078 |
+
},
|
| 2079 |
+
{
|
| 2080 |
+
"epoch": 2.1760264763677735,
|
| 2081 |
+
"grad_norm": 52.69210433959961,
|
| 2082 |
+
"learning_rate": 1.569778749922239e-06,
|
| 2083 |
+
"loss": 0.6039,
|
| 2084 |
+
"step": 1315
|
| 2085 |
+
},
|
| 2086 |
+
{
|
| 2087 |
+
"epoch": 2.1843003412969284,
|
| 2088 |
+
"grad_norm": 159.80264282226562,
|
| 2089 |
+
"learning_rate": 1.540283082291754e-06,
|
| 2090 |
+
"loss": 0.593,
|
| 2091 |
+
"step": 1320
|
| 2092 |
+
},
|
| 2093 |
+
{
|
| 2094 |
+
"epoch": 2.1925742062260833,
|
| 2095 |
+
"grad_norm": 37.81462097167969,
|
| 2096 |
+
"learning_rate": 1.5110098331177598e-06,
|
| 2097 |
+
"loss": 0.5655,
|
| 2098 |
+
"step": 1325
|
| 2099 |
+
},
|
| 2100 |
+
{
|
| 2101 |
+
"epoch": 2.2008480711552383,
|
| 2102 |
+
"grad_norm": 245.13482666015625,
|
| 2103 |
+
"learning_rate": 1.4819612022427027e-06,
|
| 2104 |
+
"loss": 0.6078,
|
| 2105 |
+
"step": 1330
|
| 2106 |
+
},
|
| 2107 |
+
{
|
| 2108 |
+
"epoch": 2.2091219360843932,
|
| 2109 |
+
"grad_norm": 139.10708618164062,
|
| 2110 |
+
"learning_rate": 1.4531393726292826e-06,
|
| 2111 |
+
"loss": 0.5248,
|
| 2112 |
+
"step": 1335
|
| 2113 |
+
},
|
| 2114 |
+
{
|
| 2115 |
+
"epoch": 2.2173958010135486,
|
| 2116 |
+
"grad_norm": 122.25520324707031,
|
| 2117 |
+
"learning_rate": 1.4245465101964164e-06,
|
| 2118 |
+
"loss": 0.5201,
|
| 2119 |
+
"step": 1340
|
| 2120 |
+
},
|
| 2121 |
+
{
|
| 2122 |
+
"epoch": 2.2256696659427035,
|
| 2123 |
+
"grad_norm": 68.17268371582031,
|
| 2124 |
+
"learning_rate": 1.3961847636564672e-06,
|
| 2125 |
+
"loss": 0.532,
|
| 2126 |
+
"step": 1345
|
| 2127 |
+
},
|
| 2128 |
+
{
|
| 2129 |
+
"epoch": 2.2339435308718585,
|
| 2130 |
+
"grad_norm": 88.55684661865234,
|
| 2131 |
+
"learning_rate": 1.3680562643537693e-06,
|
| 2132 |
+
"loss": 0.4773,
|
| 2133 |
+
"step": 1350
|
| 2134 |
+
},
|
| 2135 |
+
{
|
| 2136 |
+
"epoch": 2.2339435308718585,
|
| 2137 |
+
"eval_accuracy": 0.6874418604651162,
|
| 2138 |
+
"eval_loss": 0.6236450672149658,
|
| 2139 |
+
"eval_runtime": 139.249,
|
| 2140 |
+
"eval_samples_per_second": 7.72,
|
| 2141 |
+
"eval_steps_per_second": 7.72,
|
| 2142 |
+
"step": 1350
|
| 2143 |
+
},
|
| 2144 |
+
{
|
| 2145 |
+
"epoch": 2.2422173958010134,
|
| 2146 |
+
"grad_norm": 67.30037689208984,
|
| 2147 |
+
"learning_rate": 1.3401631261044664e-06,
|
| 2148 |
+
"loss": 0.5876,
|
| 2149 |
+
"step": 1355
|
| 2150 |
+
},
|
| 2151 |
+
{
|
| 2152 |
+
"epoch": 2.2504912607301684,
|
| 2153 |
+
"grad_norm": 66.93975067138672,
|
| 2154 |
+
"learning_rate": 1.3125074450376577e-06,
|
| 2155 |
+
"loss": 0.5292,
|
| 2156 |
+
"step": 1360
|
| 2157 |
+
},
|
| 2158 |
+
{
|
| 2159 |
+
"epoch": 2.2587651256593237,
|
| 2160 |
+
"grad_norm": 103.20861053466797,
|
| 2161 |
+
"learning_rate": 1.285091299437875e-06,
|
| 2162 |
+
"loss": 0.5353,
|
| 2163 |
+
"step": 1365
|
| 2164 |
+
},
|
| 2165 |
+
{
|
| 2166 |
+
"epoch": 2.2670389905884787,
|
| 2167 |
+
"grad_norm": 57.18837356567383,
|
| 2168 |
+
"learning_rate": 1.2579167495889117e-06,
|
| 2169 |
+
"loss": 0.525,
|
| 2170 |
+
"step": 1370
|
| 2171 |
+
},
|
| 2172 |
+
{
|
| 2173 |
+
"epoch": 2.2753128555176336,
|
| 2174 |
+
"grad_norm": 28.59246253967285,
|
| 2175 |
+
"learning_rate": 1.230985837618981e-06,
|
| 2176 |
+
"loss": 0.5938,
|
| 2177 |
+
"step": 1375
|
| 2178 |
+
},
|
| 2179 |
+
{
|
| 2180 |
+
"epoch": 2.2835867204467886,
|
| 2181 |
+
"grad_norm": 71.6416015625,
|
| 2182 |
+
"learning_rate": 1.2043005873472697e-06,
|
| 2183 |
+
"loss": 0.6092,
|
| 2184 |
+
"step": 1380
|
| 2185 |
+
},
|
| 2186 |
+
{
|
| 2187 |
+
"epoch": 2.291860585375944,
|
| 2188 |
+
"grad_norm": 89.52198028564453,
|
| 2189 |
+
"learning_rate": 1.1778630041318344e-06,
|
| 2190 |
+
"loss": 0.5926,
|
| 2191 |
+
"step": 1385
|
| 2192 |
+
},
|
| 2193 |
+
{
|
| 2194 |
+
"epoch": 2.300134450305099,
|
| 2195 |
+
"grad_norm": 119.52609252929688,
|
| 2196 |
+
"learning_rate": 1.1516750747189145e-06,
|
| 2197 |
+
"loss": 0.5102,
|
| 2198 |
+
"step": 1390
|
| 2199 |
+
},
|
| 2200 |
+
{
|
| 2201 |
+
"epoch": 2.308408315234254,
|
| 2202 |
+
"grad_norm": 27.523805618286133,
|
| 2203 |
+
"learning_rate": 1.125738767093626e-06,
|
| 2204 |
+
"loss": 0.5085,
|
| 2205 |
+
"step": 1395
|
| 2206 |
+
},
|
| 2207 |
+
{
|
| 2208 |
+
"epoch": 2.3166821801634088,
|
| 2209 |
+
"grad_norm": 108.35863494873047,
|
| 2210 |
+
"learning_rate": 1.1000560303320687e-06,
|
| 2211 |
+
"loss": 0.6225,
|
| 2212 |
+
"step": 1400
|
| 2213 |
+
},
|
| 2214 |
+
{
|
| 2215 |
+
"epoch": 2.3166821801634088,
|
| 2216 |
+
"eval_accuracy": 0.6837209302325581,
|
| 2217 |
+
"eval_loss": 0.6350362300872803,
|
| 2218 |
+
"eval_runtime": 139.1672,
|
| 2219 |
+
"eval_samples_per_second": 7.725,
|
| 2220 |
+
"eval_steps_per_second": 7.725,
|
| 2221 |
+
"step": 1400
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 2.3249560450925637,
|
| 2225 |
+
"grad_norm": 29.207040786743164,
|
| 2226 |
+
"learning_rate": 1.0746287944548576e-06,
|
| 2227 |
+
"loss": 0.5416,
|
| 2228 |
+
"step": 1405
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 2.333229910021719,
|
| 2232 |
+
"grad_norm": 67.55722045898438,
|
| 2233 |
+
"learning_rate": 1.049458970282088e-06,
|
| 2234 |
+
"loss": 0.5944,
|
| 2235 |
+
"step": 1410
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 2.341503774950874,
|
| 2239 |
+
"grad_norm": 188.1350555419922,
|
| 2240 |
+
"learning_rate": 1.0245484492897327e-06,
|
| 2241 |
+
"loss": 0.5461,
|
| 2242 |
+
"step": 1415
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 2.349777639880029,
|
| 2246 |
+
"grad_norm": 74.21896362304688,
|
| 2247 |
+
"learning_rate": 9.998991034675096e-07,
|
| 2248 |
+
"loss": 0.4793,
|
| 2249 |
+
"step": 1420
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 2.358051504809184,
|
| 2253 |
+
"grad_norm": 87.09430694580078,
|
| 2254 |
+
"learning_rate": 9.755127851781945e-07,
|
| 2255 |
+
"loss": 0.563,
|
| 2256 |
+
"step": 1425
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 2.366325369738339,
|
| 2260 |
+
"grad_norm": 236.31300354003906,
|
| 2261 |
+
"learning_rate": 9.513913270184244e-07,
|
| 2262 |
+
"loss": 0.4763,
|
| 2263 |
+
"step": 1430
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 2.3745992346674942,
|
| 2267 |
+
"grad_norm": 131.0776824951172,
|
| 2268 |
+
"learning_rate": 9.275365416809833e-07,
|
| 2269 |
+
"loss": 0.5199,
|
| 2270 |
+
"step": 1435
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 2.382873099596649,
|
| 2274 |
+
"grad_norm": 52.74057388305664,
|
| 2275 |
+
"learning_rate": 9.039502218185748e-07,
|
| 2276 |
+
"loss": 0.6553,
|
| 2277 |
+
"step": 1440
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 2.391146964525804,
|
| 2281 |
+
"grad_norm": 94.90296936035156,
|
| 2282 |
+
"learning_rate": 8.806341399091078e-07,
|
| 2283 |
+
"loss": 0.5441,
|
| 2284 |
+
"step": 1445
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 2.399420829454959,
|
| 2288 |
+
"grad_norm": 40.44979476928711,
|
| 2289 |
+
"learning_rate": 8.575900481225027e-07,
|
| 2290 |
+
"loss": 0.5655,
|
| 2291 |
+
"step": 1450
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 2.399420829454959,
|
| 2295 |
+
"eval_accuracy": 0.6893023255813954,
|
| 2296 |
+
"eval_loss": 0.6249026656150818,
|
| 2297 |
+
"eval_runtime": 135.6342,
|
| 2298 |
+
"eval_samples_per_second": 7.926,
|
| 2299 |
+
"eval_steps_per_second": 7.926,
|
| 2300 |
+
"step": 1450
|
| 2301 |
+
},
|
| 2302 |
+
{
|
| 2303 |
+
"epoch": 2.407694694384114,
|
| 2304 |
+
"grad_norm": 45.417179107666016,
|
| 2305 |
+
"learning_rate": 8.348196781890096e-07,
|
| 2306 |
+
"loss": 0.5787,
|
| 2307 |
+
"step": 1455
|
| 2308 |
+
},
|
| 2309 |
+
{
|
| 2310 |
+
"epoch": 2.4159685593132694,
|
| 2311 |
+
"grad_norm": 120.42521667480469,
|
| 2312 |
+
"learning_rate": 8.123247412690823e-07,
|
| 2313 |
+
"loss": 0.5079,
|
| 2314 |
+
"step": 1460
|
| 2315 |
+
},
|
| 2316 |
+
{
|
| 2317 |
+
"epoch": 2.4242424242424243,
|
| 2318 |
+
"grad_norm": 41.423912048339844,
|
| 2319 |
+
"learning_rate": 7.901069278247756e-07,
|
| 2320 |
+
"loss": 0.6137,
|
| 2321 |
+
"step": 1465
|
| 2322 |
+
},
|
| 2323 |
+
{
|
| 2324 |
+
"epoch": 2.4325162891715792,
|
| 2325 |
+
"grad_norm": 73.1041259765625,
|
| 2326 |
+
"learning_rate": 7.681679074927166e-07,
|
| 2327 |
+
"loss": 0.5198,
|
| 2328 |
+
"step": 1470
|
| 2329 |
+
},
|
| 2330 |
+
{
|
| 2331 |
+
"epoch": 2.440790154100734,
|
| 2332 |
+
"grad_norm": 129.06178283691406,
|
| 2333 |
+
"learning_rate": 7.465093289586345e-07,
|
| 2334 |
+
"loss": 0.5538,
|
| 2335 |
+
"step": 1475
|
| 2336 |
+
},
|
| 2337 |
+
{
|
| 2338 |
+
"epoch": 2.4490640190298896,
|
| 2339 |
+
"grad_norm": 171.7907257080078,
|
| 2340 |
+
"learning_rate": 7.25132819833459e-07,
|
| 2341 |
+
"loss": 0.5026,
|
| 2342 |
+
"step": 1480
|
| 2343 |
+
},
|
| 2344 |
+
{
|
| 2345 |
+
"epoch": 2.4573378839590445,
|
| 2346 |
+
"grad_norm": 85.45380401611328,
|
| 2347 |
+
"learning_rate": 7.04039986531011e-07,
|
| 2348 |
+
"loss": 0.6185,
|
| 2349 |
+
"step": 1485
|
| 2350 |
+
},
|
| 2351 |
+
{
|
| 2352 |
+
"epoch": 2.4656117488881995,
|
| 2353 |
+
"grad_norm": 139.45574951171875,
|
| 2354 |
+
"learning_rate": 6.832324141472841e-07,
|
| 2355 |
+
"loss": 0.5379,
|
| 2356 |
+
"step": 1490
|
| 2357 |
+
},
|
| 2358 |
+
{
|
| 2359 |
+
"epoch": 2.4738856138173544,
|
| 2360 |
+
"grad_norm": 150.1260986328125,
|
| 2361 |
+
"learning_rate": 6.627116663413249e-07,
|
| 2362 |
+
"loss": 0.5337,
|
| 2363 |
+
"step": 1495
|
| 2364 |
+
},
|
| 2365 |
+
{
|
| 2366 |
+
"epoch": 2.4821594787465093,
|
| 2367 |
+
"grad_norm": 62.410194396972656,
|
| 2368 |
+
"learning_rate": 6.424792852177275e-07,
|
| 2369 |
+
"loss": 0.5692,
|
| 2370 |
+
"step": 1500
|
| 2371 |
+
},
|
| 2372 |
+
{
|
| 2373 |
+
"epoch": 2.4821594787465093,
|
| 2374 |
+
"eval_accuracy": 0.6883720930232559,
|
| 2375 |
+
"eval_loss": 0.6257066130638123,
|
| 2376 |
+
"eval_runtime": 137.001,
|
| 2377 |
+
"eval_samples_per_second": 7.847,
|
| 2378 |
+
"eval_steps_per_second": 7.847,
|
| 2379 |
+
"step": 1500
|
| 2380 |
+
}
|
| 2381 |
+
],
|
| 2382 |
+
"logging_steps": 5,
|
| 2383 |
+
"max_steps": 1812,
|
| 2384 |
+
"num_input_tokens_seen": 0,
|
| 2385 |
+
"num_train_epochs": 3,
|
| 2386 |
+
"save_steps": 500,
|
| 2387 |
+
"total_flos": 5578659201024000.0,
|
| 2388 |
+
"train_batch_size": 1,
|
| 2389 |
+
"trial_name": null,
|
| 2390 |
+
"trial_params": null
|
| 2391 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2419f00512ba281543dd41fa6de85621e29bc67e3e2c05212bd1146dae1f3d56
|
| 3 |
+
size 5048
|