initial commit
Browse files- config.json +30 -0
- generation_config.json +7 -0
- model.safetensors +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +24 -0
- tokenizer.json +0 -0
- tokenizer_config.json +214 -0
- trainer_state.json +2373 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/home/sr2464/scratch/C2S_Files/C2S_training_runs/multicell_v2_pretraining_runs/finetuning-EleutherAI/pythia-410m-multicell_v2_pretraining-2024-07-22_18-32-23/checkpoint-12800",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"GPTNeoXForCausalLM"
|
| 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 |
+
"hidden_act": "gelu",
|
| 12 |
+
"hidden_dropout": 0.0,
|
| 13 |
+
"hidden_size": 1024,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 4096,
|
| 16 |
+
"layer_norm_eps": 1e-05,
|
| 17 |
+
"max_position_embeddings": 8192,
|
| 18 |
+
"model_type": "gpt_neox",
|
| 19 |
+
"num_attention_heads": 16,
|
| 20 |
+
"num_hidden_layers": 24,
|
| 21 |
+
"rope_scaling": null,
|
| 22 |
+
"rotary_emb_base": 10000,
|
| 23 |
+
"rotary_pct": 0.25,
|
| 24 |
+
"tie_word_embeddings": false,
|
| 25 |
+
"torch_dtype": "float32",
|
| 26 |
+
"transformers_version": "4.41.0",
|
| 27 |
+
"use_cache": false,
|
| 28 |
+
"use_parallel_residual": true,
|
| 29 |
+
"vocab_size": 50304
|
| 30 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 0,
|
| 4 |
+
"eos_token_id": 0,
|
| 5 |
+
"transformers_version": "4.41.0",
|
| 6 |
+
"use_cache": false
|
| 7 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8218b1eaff093409f1bea289cc772f2e2997546bcb7efff5a03c0965c6abbf12
|
| 3 |
+
size 1621370224
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8e2dc09cb05f68a1ed3fe17284caf797998456a00e54ac17e9ad320201709fb9
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:99c7d1aec5609095ec4a376a3918a37153fcf88bfd24238ee2bb1c11c8a92499
|
| 3 |
+
size 1064
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": "<|endoftext|>",
|
| 17 |
+
"unk_token": {
|
| 18 |
+
"content": "<|endoftext|>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": false,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
}
|
| 24 |
+
}
|
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": "<|endoftext|>",
|
| 212 |
+
"tokenizer_class": "GPTNeoXTokenizer",
|
| 213 |
+
"unk_token": "<|endoftext|>"
|
| 214 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2373 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 2.0367350578308105,
|
| 3 |
+
"best_model_checkpoint": "/home/sr2464/scratch/C2S_Files/C2S_training_runs/multicell_v2_pretraining_runs/finetuning-EleutherAI/pythia-410m-multicell_v2_pretraining-2024-07-22_18-32-23/checkpoint-15600",
|
| 4 |
+
"epoch": 0.030125854619963952,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 15600,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.00019311445269207662,
|
| 13 |
+
"grad_norm": 25.695459365844727,
|
| 14 |
+
"learning_rate": 9.655678504528514e-08,
|
| 15 |
+
"loss": 5.0491,
|
| 16 |
+
"step": 100
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.00019311445269207662,
|
| 20 |
+
"eval_loss": 4.909114360809326,
|
| 21 |
+
"eval_runtime": 453.5777,
|
| 22 |
+
"eval_samples_per_second": 4.414,
|
| 23 |
+
"eval_steps_per_second": 1.473,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.00038622890538415324,
|
| 28 |
+
"grad_norm": 9.293487548828125,
|
| 29 |
+
"learning_rate": 1.931135700905703e-07,
|
| 30 |
+
"loss": 4.7166,
|
| 31 |
+
"step": 200
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.00038622890538415324,
|
| 35 |
+
"eval_loss": 4.411260604858398,
|
| 36 |
+
"eval_runtime": 451.7078,
|
| 37 |
+
"eval_samples_per_second": 4.432,
|
| 38 |
+
"eval_steps_per_second": 1.479,
|
| 39 |
+
"step": 200
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.0005793433580762299,
|
| 43 |
+
"grad_norm": 7.290333271026611,
|
| 44 |
+
"learning_rate": 2.896703551358554e-07,
|
| 45 |
+
"loss": 4.2079,
|
| 46 |
+
"step": 300
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.0005793433580762299,
|
| 50 |
+
"eval_loss": 3.940201759338379,
|
| 51 |
+
"eval_runtime": 451.9244,
|
| 52 |
+
"eval_samples_per_second": 4.43,
|
| 53 |
+
"eval_steps_per_second": 1.478,
|
| 54 |
+
"step": 300
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.0007724578107683065,
|
| 58 |
+
"grad_norm": 4.443881511688232,
|
| 59 |
+
"learning_rate": 3.862271401811406e-07,
|
| 60 |
+
"loss": 3.8061,
|
| 61 |
+
"step": 400
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.0007724578107683065,
|
| 65 |
+
"eval_loss": 3.649780511856079,
|
| 66 |
+
"eval_runtime": 453.8369,
|
| 67 |
+
"eval_samples_per_second": 4.411,
|
| 68 |
+
"eval_steps_per_second": 1.472,
|
| 69 |
+
"step": 400
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.0009655722634603831,
|
| 73 |
+
"grad_norm": 4.046736717224121,
|
| 74 |
+
"learning_rate": 4.827839252264257e-07,
|
| 75 |
+
"loss": 3.5886,
|
| 76 |
+
"step": 500
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 0.0009655722634603831,
|
| 80 |
+
"eval_loss": 3.4819984436035156,
|
| 81 |
+
"eval_runtime": 454.1608,
|
| 82 |
+
"eval_samples_per_second": 4.408,
|
| 83 |
+
"eval_steps_per_second": 1.471,
|
| 84 |
+
"step": 500
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.0011586867161524597,
|
| 88 |
+
"grad_norm": 4.199274063110352,
|
| 89 |
+
"learning_rate": 5.793407102717108e-07,
|
| 90 |
+
"loss": 3.4235,
|
| 91 |
+
"step": 600
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.0011586867161524597,
|
| 95 |
+
"eval_loss": 3.3766062259674072,
|
| 96 |
+
"eval_runtime": 451.2045,
|
| 97 |
+
"eval_samples_per_second": 4.437,
|
| 98 |
+
"eval_steps_per_second": 1.48,
|
| 99 |
+
"step": 600
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.0013518011688445364,
|
| 103 |
+
"grad_norm": 3.957104444503784,
|
| 104 |
+
"learning_rate": 6.75897495316996e-07,
|
| 105 |
+
"loss": 3.3368,
|
| 106 |
+
"step": 700
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.0013518011688445364,
|
| 110 |
+
"eval_loss": 3.298933982849121,
|
| 111 |
+
"eval_runtime": 452.0648,
|
| 112 |
+
"eval_samples_per_second": 4.429,
|
| 113 |
+
"eval_steps_per_second": 1.478,
|
| 114 |
+
"step": 700
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.001544915621536613,
|
| 118 |
+
"grad_norm": 3.570394992828369,
|
| 119 |
+
"learning_rate": 7.724542803622812e-07,
|
| 120 |
+
"loss": 3.258,
|
| 121 |
+
"step": 800
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.001544915621536613,
|
| 125 |
+
"eval_loss": 3.2292919158935547,
|
| 126 |
+
"eval_runtime": 455.0475,
|
| 127 |
+
"eval_samples_per_second": 4.4,
|
| 128 |
+
"eval_steps_per_second": 1.468,
|
| 129 |
+
"step": 800
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.0017380300742286896,
|
| 133 |
+
"grad_norm": 2.5731029510498047,
|
| 134 |
+
"learning_rate": 8.690110654075663e-07,
|
| 135 |
+
"loss": 3.1963,
|
| 136 |
+
"step": 900
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.0017380300742286896,
|
| 140 |
+
"eval_loss": 3.1713345050811768,
|
| 141 |
+
"eval_runtime": 451.7557,
|
| 142 |
+
"eval_samples_per_second": 4.432,
|
| 143 |
+
"eval_steps_per_second": 1.479,
|
| 144 |
+
"step": 900
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.0019311445269207662,
|
| 148 |
+
"grad_norm": 4.118990898132324,
|
| 149 |
+
"learning_rate": 9.655678504528513e-07,
|
| 150 |
+
"loss": 3.146,
|
| 151 |
+
"step": 1000
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.0019311445269207662,
|
| 155 |
+
"eval_loss": 3.1244723796844482,
|
| 156 |
+
"eval_runtime": 445.9466,
|
| 157 |
+
"eval_samples_per_second": 4.489,
|
| 158 |
+
"eval_steps_per_second": 1.498,
|
| 159 |
+
"step": 1000
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.002124258979612843,
|
| 163 |
+
"grad_norm": 3.6328647136688232,
|
| 164 |
+
"learning_rate": 1.0621246354981365e-06,
|
| 165 |
+
"loss": 3.1092,
|
| 166 |
+
"step": 1100
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.002124258979612843,
|
| 170 |
+
"eval_loss": 3.0821051597595215,
|
| 171 |
+
"eval_runtime": 445.7792,
|
| 172 |
+
"eval_samples_per_second": 4.491,
|
| 173 |
+
"eval_steps_per_second": 1.498,
|
| 174 |
+
"step": 1100
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.0023173734323049195,
|
| 178 |
+
"grad_norm": 3.1631813049316406,
|
| 179 |
+
"learning_rate": 1.1586814205434216e-06,
|
| 180 |
+
"loss": 3.0401,
|
| 181 |
+
"step": 1200
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.0023173734323049195,
|
| 185 |
+
"eval_loss": 3.04966402053833,
|
| 186 |
+
"eval_runtime": 445.2862,
|
| 187 |
+
"eval_samples_per_second": 4.496,
|
| 188 |
+
"eval_steps_per_second": 1.5,
|
| 189 |
+
"step": 1200
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.0025104878849969963,
|
| 193 |
+
"grad_norm": 3.6014091968536377,
|
| 194 |
+
"learning_rate": 1.2552382055887068e-06,
|
| 195 |
+
"loss": 3.0162,
|
| 196 |
+
"step": 1300
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 0.0025104878849969963,
|
| 200 |
+
"eval_loss": 3.0158376693725586,
|
| 201 |
+
"eval_runtime": 445.2448,
|
| 202 |
+
"eval_samples_per_second": 4.496,
|
| 203 |
+
"eval_steps_per_second": 1.5,
|
| 204 |
+
"step": 1300
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.0027036023376890727,
|
| 208 |
+
"grad_norm": 3.6174280643463135,
|
| 209 |
+
"learning_rate": 1.351794990633992e-06,
|
| 210 |
+
"loss": 2.9725,
|
| 211 |
+
"step": 1400
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 0.0027036023376890727,
|
| 215 |
+
"eval_loss": 2.979250431060791,
|
| 216 |
+
"eval_runtime": 448.3628,
|
| 217 |
+
"eval_samples_per_second": 4.465,
|
| 218 |
+
"eval_steps_per_second": 1.49,
|
| 219 |
+
"step": 1400
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.0028967167903811496,
|
| 223 |
+
"grad_norm": 4.053787708282471,
|
| 224 |
+
"learning_rate": 1.448351775679277e-06,
|
| 225 |
+
"loss": 2.9443,
|
| 226 |
+
"step": 1500
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.0028967167903811496,
|
| 230 |
+
"eval_loss": 2.938361883163452,
|
| 231 |
+
"eval_runtime": 445.077,
|
| 232 |
+
"eval_samples_per_second": 4.498,
|
| 233 |
+
"eval_steps_per_second": 1.501,
|
| 234 |
+
"step": 1500
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.003089831243073226,
|
| 238 |
+
"grad_norm": 4.192756652832031,
|
| 239 |
+
"learning_rate": 1.5449085607245623e-06,
|
| 240 |
+
"loss": 2.9164,
|
| 241 |
+
"step": 1600
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.003089831243073226,
|
| 245 |
+
"eval_loss": 2.89266037940979,
|
| 246 |
+
"eval_runtime": 474.6898,
|
| 247 |
+
"eval_samples_per_second": 4.217,
|
| 248 |
+
"eval_steps_per_second": 1.407,
|
| 249 |
+
"step": 1600
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.003282945695765303,
|
| 253 |
+
"grad_norm": 4.469438076019287,
|
| 254 |
+
"learning_rate": 1.6414653457698475e-06,
|
| 255 |
+
"loss": 2.8643,
|
| 256 |
+
"step": 1700
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 0.003282945695765303,
|
| 260 |
+
"eval_loss": 2.851994752883911,
|
| 261 |
+
"eval_runtime": 470.3677,
|
| 262 |
+
"eval_samples_per_second": 4.256,
|
| 263 |
+
"eval_steps_per_second": 1.42,
|
| 264 |
+
"step": 1700
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.003476060148457379,
|
| 268 |
+
"grad_norm": 4.589786052703857,
|
| 269 |
+
"learning_rate": 1.7380221308151326e-06,
|
| 270 |
+
"loss": 2.8158,
|
| 271 |
+
"step": 1800
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 0.003476060148457379,
|
| 275 |
+
"eval_loss": 2.8121631145477295,
|
| 276 |
+
"eval_runtime": 474.8649,
|
| 277 |
+
"eval_samples_per_second": 4.216,
|
| 278 |
+
"eval_steps_per_second": 1.407,
|
| 279 |
+
"step": 1800
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.003669174601149456,
|
| 283 |
+
"grad_norm": 4.862005710601807,
|
| 284 |
+
"learning_rate": 1.8345789158604178e-06,
|
| 285 |
+
"loss": 2.7741,
|
| 286 |
+
"step": 1900
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 0.003669174601149456,
|
| 290 |
+
"eval_loss": 2.767763614654541,
|
| 291 |
+
"eval_runtime": 451.5831,
|
| 292 |
+
"eval_samples_per_second": 4.433,
|
| 293 |
+
"eval_steps_per_second": 1.479,
|
| 294 |
+
"step": 1900
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 0.0038622890538415324,
|
| 298 |
+
"grad_norm": 8.818233489990234,
|
| 299 |
+
"learning_rate": 1.9311357009057026e-06,
|
| 300 |
+
"loss": 2.7214,
|
| 301 |
+
"step": 2000
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 0.0038622890538415324,
|
| 305 |
+
"eval_loss": 2.727350950241089,
|
| 306 |
+
"eval_runtime": 450.1021,
|
| 307 |
+
"eval_samples_per_second": 4.448,
|
| 308 |
+
"eval_steps_per_second": 1.484,
|
| 309 |
+
"step": 2000
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.004055403506533609,
|
| 313 |
+
"grad_norm": 6.361968040466309,
|
| 314 |
+
"learning_rate": 2.027692485950988e-06,
|
| 315 |
+
"loss": 2.6898,
|
| 316 |
+
"step": 2100
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 0.004055403506533609,
|
| 320 |
+
"eval_loss": 2.6967461109161377,
|
| 321 |
+
"eval_runtime": 455.1751,
|
| 322 |
+
"eval_samples_per_second": 4.398,
|
| 323 |
+
"eval_steps_per_second": 1.468,
|
| 324 |
+
"step": 2100
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.004248517959225686,
|
| 328 |
+
"grad_norm": 3.338510513305664,
|
| 329 |
+
"learning_rate": 2.124249270996273e-06,
|
| 330 |
+
"loss": 2.6505,
|
| 331 |
+
"step": 2200
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.004248517959225686,
|
| 335 |
+
"eval_loss": 2.6587347984313965,
|
| 336 |
+
"eval_runtime": 454.0116,
|
| 337 |
+
"eval_samples_per_second": 4.41,
|
| 338 |
+
"eval_steps_per_second": 1.471,
|
| 339 |
+
"step": 2200
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.004441632411917762,
|
| 343 |
+
"grad_norm": 3.8161659240722656,
|
| 344 |
+
"learning_rate": 2.2208060560415583e-06,
|
| 345 |
+
"loss": 2.6135,
|
| 346 |
+
"step": 2300
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.004441632411917762,
|
| 350 |
+
"eval_loss": 2.6250338554382324,
|
| 351 |
+
"eval_runtime": 455.8437,
|
| 352 |
+
"eval_samples_per_second": 4.392,
|
| 353 |
+
"eval_steps_per_second": 1.465,
|
| 354 |
+
"step": 2300
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.004634746864609839,
|
| 358 |
+
"grad_norm": 4.693429946899414,
|
| 359 |
+
"learning_rate": 2.317362841086843e-06,
|
| 360 |
+
"loss": 2.5805,
|
| 361 |
+
"step": 2400
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.004634746864609839,
|
| 365 |
+
"eval_loss": 2.5991430282592773,
|
| 366 |
+
"eval_runtime": 450.2467,
|
| 367 |
+
"eval_samples_per_second": 4.446,
|
| 368 |
+
"eval_steps_per_second": 1.484,
|
| 369 |
+
"step": 2400
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.004827861317301916,
|
| 373 |
+
"grad_norm": 3.4596662521362305,
|
| 374 |
+
"learning_rate": 2.4139196261321284e-06,
|
| 375 |
+
"loss": 2.5585,
|
| 376 |
+
"step": 2500
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 0.004827861317301916,
|
| 380 |
+
"eval_loss": 2.563077926635742,
|
| 381 |
+
"eval_runtime": 454.4394,
|
| 382 |
+
"eval_samples_per_second": 4.405,
|
| 383 |
+
"eval_steps_per_second": 1.47,
|
| 384 |
+
"step": 2500
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.005020975769993993,
|
| 388 |
+
"grad_norm": 5.340704917907715,
|
| 389 |
+
"learning_rate": 2.5104764111774136e-06,
|
| 390 |
+
"loss": 2.5134,
|
| 391 |
+
"step": 2600
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 0.005020975769993993,
|
| 395 |
+
"eval_loss": 2.5388693809509277,
|
| 396 |
+
"eval_runtime": 454.2794,
|
| 397 |
+
"eval_samples_per_second": 4.407,
|
| 398 |
+
"eval_steps_per_second": 1.47,
|
| 399 |
+
"step": 2600
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 0.005214090222686069,
|
| 403 |
+
"grad_norm": 3.7562334537506104,
|
| 404 |
+
"learning_rate": 2.6070331962226984e-06,
|
| 405 |
+
"loss": 2.4956,
|
| 406 |
+
"step": 2700
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 0.005214090222686069,
|
| 410 |
+
"eval_loss": 2.5175445079803467,
|
| 411 |
+
"eval_runtime": 454.0268,
|
| 412 |
+
"eval_samples_per_second": 4.409,
|
| 413 |
+
"eval_steps_per_second": 1.471,
|
| 414 |
+
"step": 2700
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 0.005407204675378145,
|
| 418 |
+
"grad_norm": 4.657394886016846,
|
| 419 |
+
"learning_rate": 2.703589981267984e-06,
|
| 420 |
+
"loss": 2.484,
|
| 421 |
+
"step": 2800
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 0.005407204675378145,
|
| 425 |
+
"eval_loss": 2.493380069732666,
|
| 426 |
+
"eval_runtime": 451.8282,
|
| 427 |
+
"eval_samples_per_second": 4.431,
|
| 428 |
+
"eval_steps_per_second": 1.478,
|
| 429 |
+
"step": 2800
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.005600319128070222,
|
| 433 |
+
"grad_norm": 4.6835150718688965,
|
| 434 |
+
"learning_rate": 2.800146766313269e-06,
|
| 435 |
+
"loss": 2.4705,
|
| 436 |
+
"step": 2900
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.005600319128070222,
|
| 440 |
+
"eval_loss": 2.4784457683563232,
|
| 441 |
+
"eval_runtime": 451.6775,
|
| 442 |
+
"eval_samples_per_second": 4.432,
|
| 443 |
+
"eval_steps_per_second": 1.479,
|
| 444 |
+
"step": 2900
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.005793433580762299,
|
| 448 |
+
"grad_norm": 5.81275749206543,
|
| 449 |
+
"learning_rate": 2.896703551358554e-06,
|
| 450 |
+
"loss": 2.4405,
|
| 451 |
+
"step": 3000
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.005793433580762299,
|
| 455 |
+
"eval_loss": 2.4617481231689453,
|
| 456 |
+
"eval_runtime": 453.6192,
|
| 457 |
+
"eval_samples_per_second": 4.413,
|
| 458 |
+
"eval_steps_per_second": 1.473,
|
| 459 |
+
"step": 3000
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 0.005986548033454375,
|
| 463 |
+
"grad_norm": 3.0662126541137695,
|
| 464 |
+
"learning_rate": 2.9932603364038394e-06,
|
| 465 |
+
"loss": 2.427,
|
| 466 |
+
"step": 3100
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 0.005986548033454375,
|
| 470 |
+
"eval_loss": 2.448239326477051,
|
| 471 |
+
"eval_runtime": 451.3577,
|
| 472 |
+
"eval_samples_per_second": 4.436,
|
| 473 |
+
"eval_steps_per_second": 1.48,
|
| 474 |
+
"step": 3100
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 0.006179662486146452,
|
| 478 |
+
"grad_norm": 7.125177383422852,
|
| 479 |
+
"learning_rate": 3.0898171214491246e-06,
|
| 480 |
+
"loss": 2.41,
|
| 481 |
+
"step": 3200
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 0.006179662486146452,
|
| 485 |
+
"eval_loss": 2.435983896255493,
|
| 486 |
+
"eval_runtime": 451.9197,
|
| 487 |
+
"eval_samples_per_second": 4.43,
|
| 488 |
+
"eval_steps_per_second": 1.478,
|
| 489 |
+
"step": 3200
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 0.006372776938838529,
|
| 493 |
+
"grad_norm": 5.4418439865112305,
|
| 494 |
+
"learning_rate": 3.1863739064944094e-06,
|
| 495 |
+
"loss": 2.401,
|
| 496 |
+
"step": 3300
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 0.006372776938838529,
|
| 500 |
+
"eval_loss": 2.42331862449646,
|
| 501 |
+
"eval_runtime": 449.2333,
|
| 502 |
+
"eval_samples_per_second": 4.456,
|
| 503 |
+
"eval_steps_per_second": 1.487,
|
| 504 |
+
"step": 3300
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 0.006565891391530606,
|
| 508 |
+
"grad_norm": 10.877897262573242,
|
| 509 |
+
"learning_rate": 3.282930691539695e-06,
|
| 510 |
+
"loss": 2.3891,
|
| 511 |
+
"step": 3400
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 0.006565891391530606,
|
| 515 |
+
"eval_loss": 2.417658567428589,
|
| 516 |
+
"eval_runtime": 453.4294,
|
| 517 |
+
"eval_samples_per_second": 4.415,
|
| 518 |
+
"eval_steps_per_second": 1.473,
|
| 519 |
+
"step": 3400
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 0.006759005844222682,
|
| 523 |
+
"grad_norm": 5.651215076446533,
|
| 524 |
+
"learning_rate": 3.37948747658498e-06,
|
| 525 |
+
"loss": 2.3852,
|
| 526 |
+
"step": 3500
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 0.006759005844222682,
|
| 530 |
+
"eval_loss": 2.4044547080993652,
|
| 531 |
+
"eval_runtime": 456.5451,
|
| 532 |
+
"eval_samples_per_second": 4.385,
|
| 533 |
+
"eval_steps_per_second": 1.463,
|
| 534 |
+
"step": 3500
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.006952120296914758,
|
| 538 |
+
"grad_norm": 4.415964126586914,
|
| 539 |
+
"learning_rate": 3.476044261630265e-06,
|
| 540 |
+
"loss": 2.3623,
|
| 541 |
+
"step": 3600
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.006952120296914758,
|
| 545 |
+
"eval_loss": 2.3921079635620117,
|
| 546 |
+
"eval_runtime": 455.241,
|
| 547 |
+
"eval_samples_per_second": 4.398,
|
| 548 |
+
"eval_steps_per_second": 1.467,
|
| 549 |
+
"step": 3600
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.007145234749606835,
|
| 553 |
+
"grad_norm": 4.4306793212890625,
|
| 554 |
+
"learning_rate": 3.57260104667555e-06,
|
| 555 |
+
"loss": 2.3519,
|
| 556 |
+
"step": 3700
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.007145234749606835,
|
| 560 |
+
"eval_loss": 2.3823041915893555,
|
| 561 |
+
"eval_runtime": 453.1329,
|
| 562 |
+
"eval_samples_per_second": 4.418,
|
| 563 |
+
"eval_steps_per_second": 1.474,
|
| 564 |
+
"step": 3700
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 0.007338349202298912,
|
| 568 |
+
"grad_norm": 4.322727203369141,
|
| 569 |
+
"learning_rate": 3.6691578317208356e-06,
|
| 570 |
+
"loss": 2.3449,
|
| 571 |
+
"step": 3800
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 0.007338349202298912,
|
| 575 |
+
"eval_loss": 2.3733458518981934,
|
| 576 |
+
"eval_runtime": 452.5517,
|
| 577 |
+
"eval_samples_per_second": 4.424,
|
| 578 |
+
"eval_steps_per_second": 1.476,
|
| 579 |
+
"step": 3800
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 0.007531463654990988,
|
| 583 |
+
"grad_norm": 3.902733325958252,
|
| 584 |
+
"learning_rate": 3.76571461676612e-06,
|
| 585 |
+
"loss": 2.3425,
|
| 586 |
+
"step": 3900
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 0.007531463654990988,
|
| 590 |
+
"eval_loss": 2.3656857013702393,
|
| 591 |
+
"eval_runtime": 470.1556,
|
| 592 |
+
"eval_samples_per_second": 4.258,
|
| 593 |
+
"eval_steps_per_second": 1.421,
|
| 594 |
+
"step": 3900
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 0.007724578107683065,
|
| 598 |
+
"grad_norm": 8.922844886779785,
|
| 599 |
+
"learning_rate": 3.862271401811405e-06,
|
| 600 |
+
"loss": 2.3372,
|
| 601 |
+
"step": 4000
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 0.007724578107683065,
|
| 605 |
+
"eval_loss": 2.3647260665893555,
|
| 606 |
+
"eval_runtime": 473.4275,
|
| 607 |
+
"eval_samples_per_second": 4.229,
|
| 608 |
+
"eval_steps_per_second": 1.411,
|
| 609 |
+
"step": 4000
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 0.00791769256037514,
|
| 613 |
+
"grad_norm": 5.44854736328125,
|
| 614 |
+
"learning_rate": 3.9588281868566905e-06,
|
| 615 |
+
"loss": 2.3265,
|
| 616 |
+
"step": 4100
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 0.00791769256037514,
|
| 620 |
+
"eval_loss": 2.354055643081665,
|
| 621 |
+
"eval_runtime": 469.6055,
|
| 622 |
+
"eval_samples_per_second": 4.263,
|
| 623 |
+
"eval_steps_per_second": 1.422,
|
| 624 |
+
"step": 4100
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 0.008110807013067219,
|
| 628 |
+
"grad_norm": 3.362586736679077,
|
| 629 |
+
"learning_rate": 4.055384971901976e-06,
|
| 630 |
+
"loss": 2.3184,
|
| 631 |
+
"step": 4200
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 0.008110807013067219,
|
| 635 |
+
"eval_loss": 2.3412978649139404,
|
| 636 |
+
"eval_runtime": 475.6279,
|
| 637 |
+
"eval_samples_per_second": 4.209,
|
| 638 |
+
"eval_steps_per_second": 1.404,
|
| 639 |
+
"step": 4200
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.008303921465759295,
|
| 643 |
+
"grad_norm": 2.9980242252349854,
|
| 644 |
+
"learning_rate": 4.151941756947261e-06,
|
| 645 |
+
"loss": 2.3133,
|
| 646 |
+
"step": 4300
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.008303921465759295,
|
| 650 |
+
"eval_loss": 2.3344852924346924,
|
| 651 |
+
"eval_runtime": 471.0361,
|
| 652 |
+
"eval_samples_per_second": 4.25,
|
| 653 |
+
"eval_steps_per_second": 1.418,
|
| 654 |
+
"step": 4300
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.008497035918451372,
|
| 658 |
+
"grad_norm": 2.9420857429504395,
|
| 659 |
+
"learning_rate": 4.248498541992546e-06,
|
| 660 |
+
"loss": 2.298,
|
| 661 |
+
"step": 4400
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.008497035918451372,
|
| 665 |
+
"eval_loss": 2.3255553245544434,
|
| 666 |
+
"eval_runtime": 469.2896,
|
| 667 |
+
"eval_samples_per_second": 4.266,
|
| 668 |
+
"eval_steps_per_second": 1.423,
|
| 669 |
+
"step": 4400
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 0.008690150371143448,
|
| 673 |
+
"grad_norm": 3.5292468070983887,
|
| 674 |
+
"learning_rate": 4.345055327037831e-06,
|
| 675 |
+
"loss": 2.2892,
|
| 676 |
+
"step": 4500
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 0.008690150371143448,
|
| 680 |
+
"eval_loss": 2.3173961639404297,
|
| 681 |
+
"eval_runtime": 446.9797,
|
| 682 |
+
"eval_samples_per_second": 4.479,
|
| 683 |
+
"eval_steps_per_second": 1.494,
|
| 684 |
+
"step": 4500
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 0.008883264823835524,
|
| 688 |
+
"grad_norm": 3.7100796699523926,
|
| 689 |
+
"learning_rate": 4.441612112083117e-06,
|
| 690 |
+
"loss": 2.2829,
|
| 691 |
+
"step": 4600
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 0.008883264823835524,
|
| 695 |
+
"eval_loss": 2.313410520553589,
|
| 696 |
+
"eval_runtime": 443.1163,
|
| 697 |
+
"eval_samples_per_second": 4.518,
|
| 698 |
+
"eval_steps_per_second": 1.508,
|
| 699 |
+
"step": 4600
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 0.009076379276527602,
|
| 703 |
+
"grad_norm": 3.060458183288574,
|
| 704 |
+
"learning_rate": 4.538168897128402e-06,
|
| 705 |
+
"loss": 2.2622,
|
| 706 |
+
"step": 4700
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 0.009076379276527602,
|
| 710 |
+
"eval_loss": 2.303041934967041,
|
| 711 |
+
"eval_runtime": 515.3629,
|
| 712 |
+
"eval_samples_per_second": 3.885,
|
| 713 |
+
"eval_steps_per_second": 1.296,
|
| 714 |
+
"step": 4700
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 0.009269493729219678,
|
| 718 |
+
"grad_norm": 2.960869312286377,
|
| 719 |
+
"learning_rate": 4.634725682173686e-06,
|
| 720 |
+
"loss": 2.2691,
|
| 721 |
+
"step": 4800
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 0.009269493729219678,
|
| 725 |
+
"eval_loss": 2.2965316772460938,
|
| 726 |
+
"eval_runtime": 519.7712,
|
| 727 |
+
"eval_samples_per_second": 3.852,
|
| 728 |
+
"eval_steps_per_second": 1.285,
|
| 729 |
+
"step": 4800
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 0.009462608181911754,
|
| 733 |
+
"grad_norm": 4.1272993087768555,
|
| 734 |
+
"learning_rate": 4.7312824672189715e-06,
|
| 735 |
+
"loss": 2.2662,
|
| 736 |
+
"step": 4900
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 0.009462608181911754,
|
| 740 |
+
"eval_loss": 2.2936575412750244,
|
| 741 |
+
"eval_runtime": 517.9358,
|
| 742 |
+
"eval_samples_per_second": 3.865,
|
| 743 |
+
"eval_steps_per_second": 1.29,
|
| 744 |
+
"step": 4900
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.009655722634603832,
|
| 748 |
+
"grad_norm": 3.13708758354187,
|
| 749 |
+
"learning_rate": 4.827839252264257e-06,
|
| 750 |
+
"loss": 2.2566,
|
| 751 |
+
"step": 5000
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.009655722634603832,
|
| 755 |
+
"eval_loss": 2.282074451446533,
|
| 756 |
+
"eval_runtime": 518.3534,
|
| 757 |
+
"eval_samples_per_second": 3.862,
|
| 758 |
+
"eval_steps_per_second": 1.289,
|
| 759 |
+
"step": 5000
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.009848837087295908,
|
| 763 |
+
"grad_norm": 6.472213268280029,
|
| 764 |
+
"learning_rate": 4.924396037309542e-06,
|
| 765 |
+
"loss": 2.2491,
|
| 766 |
+
"step": 5100
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.009848837087295908,
|
| 770 |
+
"eval_loss": 2.2818562984466553,
|
| 771 |
+
"eval_runtime": 516.8354,
|
| 772 |
+
"eval_samples_per_second": 3.874,
|
| 773 |
+
"eval_steps_per_second": 1.292,
|
| 774 |
+
"step": 5100
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 0.010041951539987985,
|
| 778 |
+
"grad_norm": 3.39032244682312,
|
| 779 |
+
"learning_rate": 5.020952822354827e-06,
|
| 780 |
+
"loss": 2.2439,
|
| 781 |
+
"step": 5200
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 0.010041951539987985,
|
| 785 |
+
"eval_loss": 2.2728891372680664,
|
| 786 |
+
"eval_runtime": 519.2594,
|
| 787 |
+
"eval_samples_per_second": 3.855,
|
| 788 |
+
"eval_steps_per_second": 1.286,
|
| 789 |
+
"step": 5200
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 0.010235065992680061,
|
| 793 |
+
"grad_norm": 2.733482837677002,
|
| 794 |
+
"learning_rate": 5.1175096074001125e-06,
|
| 795 |
+
"loss": 2.2377,
|
| 796 |
+
"step": 5300
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 0.010235065992680061,
|
| 800 |
+
"eval_loss": 2.266763687133789,
|
| 801 |
+
"eval_runtime": 516.9841,
|
| 802 |
+
"eval_samples_per_second": 3.872,
|
| 803 |
+
"eval_steps_per_second": 1.292,
|
| 804 |
+
"step": 5300
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 0.010428180445372137,
|
| 808 |
+
"grad_norm": 3.014375686645508,
|
| 809 |
+
"learning_rate": 5.214066392445397e-06,
|
| 810 |
+
"loss": 2.229,
|
| 811 |
+
"step": 5400
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 0.010428180445372137,
|
| 815 |
+
"eval_loss": 2.2636349201202393,
|
| 816 |
+
"eval_runtime": 518.5144,
|
| 817 |
+
"eval_samples_per_second": 3.861,
|
| 818 |
+
"eval_steps_per_second": 1.288,
|
| 819 |
+
"step": 5400
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 0.010621294898064215,
|
| 823 |
+
"grad_norm": 3.083723306655884,
|
| 824 |
+
"learning_rate": 5.310623177490683e-06,
|
| 825 |
+
"loss": 2.2328,
|
| 826 |
+
"step": 5500
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 0.010621294898064215,
|
| 830 |
+
"eval_loss": 2.258470058441162,
|
| 831 |
+
"eval_runtime": 516.2515,
|
| 832 |
+
"eval_samples_per_second": 3.878,
|
| 833 |
+
"eval_steps_per_second": 1.294,
|
| 834 |
+
"step": 5500
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 0.01081440935075629,
|
| 838 |
+
"grad_norm": 2.823312997817993,
|
| 839 |
+
"learning_rate": 5.407179962535968e-06,
|
| 840 |
+
"loss": 2.2203,
|
| 841 |
+
"step": 5600
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 0.01081440935075629,
|
| 845 |
+
"eval_loss": 2.2555058002471924,
|
| 846 |
+
"eval_runtime": 517.1852,
|
| 847 |
+
"eval_samples_per_second": 3.871,
|
| 848 |
+
"eval_steps_per_second": 1.292,
|
| 849 |
+
"step": 5600
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.011007523803448367,
|
| 853 |
+
"grad_norm": 2.4301838874816895,
|
| 854 |
+
"learning_rate": 5.5037367475812526e-06,
|
| 855 |
+
"loss": 2.219,
|
| 856 |
+
"step": 5700
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.011007523803448367,
|
| 860 |
+
"eval_loss": 2.2486064434051514,
|
| 861 |
+
"eval_runtime": 519.4056,
|
| 862 |
+
"eval_samples_per_second": 3.854,
|
| 863 |
+
"eval_steps_per_second": 1.286,
|
| 864 |
+
"step": 5700
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.011200638256140445,
|
| 868 |
+
"grad_norm": 2.8203349113464355,
|
| 869 |
+
"learning_rate": 5.600293532626538e-06,
|
| 870 |
+
"loss": 2.2194,
|
| 871 |
+
"step": 5800
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.011200638256140445,
|
| 875 |
+
"eval_loss": 2.2447359561920166,
|
| 876 |
+
"eval_runtime": 517.18,
|
| 877 |
+
"eval_samples_per_second": 3.871,
|
| 878 |
+
"eval_steps_per_second": 1.292,
|
| 879 |
+
"step": 5800
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 0.01139375270883252,
|
| 883 |
+
"grad_norm": 3.018662214279175,
|
| 884 |
+
"learning_rate": 5.696850317671823e-06,
|
| 885 |
+
"loss": 2.216,
|
| 886 |
+
"step": 5900
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 0.01139375270883252,
|
| 890 |
+
"eval_loss": 2.2398221492767334,
|
| 891 |
+
"eval_runtime": 526.6953,
|
| 892 |
+
"eval_samples_per_second": 3.801,
|
| 893 |
+
"eval_steps_per_second": 1.268,
|
| 894 |
+
"step": 5900
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 0.011586867161524598,
|
| 898 |
+
"grad_norm": 3.5099599361419678,
|
| 899 |
+
"learning_rate": 5.793407102717108e-06,
|
| 900 |
+
"loss": 2.2066,
|
| 901 |
+
"step": 6000
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 0.011586867161524598,
|
| 905 |
+
"eval_loss": 2.2389719486236572,
|
| 906 |
+
"eval_runtime": 454.8275,
|
| 907 |
+
"eval_samples_per_second": 4.402,
|
| 908 |
+
"eval_steps_per_second": 1.469,
|
| 909 |
+
"step": 6000
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 0.011779981614216674,
|
| 913 |
+
"grad_norm": 2.642719268798828,
|
| 914 |
+
"learning_rate": 5.8899638877623935e-06,
|
| 915 |
+
"loss": 2.199,
|
| 916 |
+
"step": 6100
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 0.011779981614216674,
|
| 920 |
+
"eval_loss": 2.2292520999908447,
|
| 921 |
+
"eval_runtime": 453.5237,
|
| 922 |
+
"eval_samples_per_second": 4.414,
|
| 923 |
+
"eval_steps_per_second": 1.473,
|
| 924 |
+
"step": 6100
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 0.01197309606690875,
|
| 928 |
+
"grad_norm": 2.63216233253479,
|
| 929 |
+
"learning_rate": 5.986520672807679e-06,
|
| 930 |
+
"loss": 2.1942,
|
| 931 |
+
"step": 6200
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 0.01197309606690875,
|
| 935 |
+
"eval_loss": 2.224433183670044,
|
| 936 |
+
"eval_runtime": 454.9765,
|
| 937 |
+
"eval_samples_per_second": 4.4,
|
| 938 |
+
"eval_steps_per_second": 1.468,
|
| 939 |
+
"step": 6200
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 0.012166210519600828,
|
| 943 |
+
"grad_norm": 3.0821280479431152,
|
| 944 |
+
"learning_rate": 6.083077457852963e-06,
|
| 945 |
+
"loss": 2.1926,
|
| 946 |
+
"step": 6300
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 0.012166210519600828,
|
| 950 |
+
"eval_loss": 2.2213406562805176,
|
| 951 |
+
"eval_runtime": 455.9375,
|
| 952 |
+
"eval_samples_per_second": 4.391,
|
| 953 |
+
"eval_steps_per_second": 1.465,
|
| 954 |
+
"step": 6300
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.012359324972292904,
|
| 958 |
+
"grad_norm": 3.2778172492980957,
|
| 959 |
+
"learning_rate": 6.179634242898249e-06,
|
| 960 |
+
"loss": 2.1771,
|
| 961 |
+
"step": 6400
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.012359324972292904,
|
| 965 |
+
"eval_loss": 2.221287488937378,
|
| 966 |
+
"eval_runtime": 451.8099,
|
| 967 |
+
"eval_samples_per_second": 4.431,
|
| 968 |
+
"eval_steps_per_second": 1.478,
|
| 969 |
+
"step": 6400
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.01255243942498498,
|
| 973 |
+
"grad_norm": 1.9902693033218384,
|
| 974 |
+
"learning_rate": 6.2761910279435345e-06,
|
| 975 |
+
"loss": 2.1654,
|
| 976 |
+
"step": 6500
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.01255243942498498,
|
| 980 |
+
"eval_loss": 2.2130887508392334,
|
| 981 |
+
"eval_runtime": 452.4518,
|
| 982 |
+
"eval_samples_per_second": 4.425,
|
| 983 |
+
"eval_steps_per_second": 1.476,
|
| 984 |
+
"step": 6500
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 0.012745553877677058,
|
| 988 |
+
"grad_norm": 3.490743637084961,
|
| 989 |
+
"learning_rate": 6.372747812988819e-06,
|
| 990 |
+
"loss": 2.1891,
|
| 991 |
+
"step": 6600
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 0.012745553877677058,
|
| 995 |
+
"eval_loss": 2.210174083709717,
|
| 996 |
+
"eval_runtime": 481.3251,
|
| 997 |
+
"eval_samples_per_second": 4.159,
|
| 998 |
+
"eval_steps_per_second": 1.388,
|
| 999 |
+
"step": 6600
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 0.012938668330369133,
|
| 1003 |
+
"grad_norm": 2.3788084983825684,
|
| 1004 |
+
"learning_rate": 6.469304598034104e-06,
|
| 1005 |
+
"loss": 2.1834,
|
| 1006 |
+
"step": 6700
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 0.012938668330369133,
|
| 1010 |
+
"eval_loss": 2.2062604427337646,
|
| 1011 |
+
"eval_runtime": 452.5564,
|
| 1012 |
+
"eval_samples_per_second": 4.424,
|
| 1013 |
+
"eval_steps_per_second": 1.476,
|
| 1014 |
+
"step": 6700
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 0.013131782783061211,
|
| 1018 |
+
"grad_norm": 2.141946792602539,
|
| 1019 |
+
"learning_rate": 6.56586138307939e-06,
|
| 1020 |
+
"loss": 2.1605,
|
| 1021 |
+
"step": 6800
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"epoch": 0.013131782783061211,
|
| 1025 |
+
"eval_loss": 2.2010421752929688,
|
| 1026 |
+
"eval_runtime": 455.2341,
|
| 1027 |
+
"eval_samples_per_second": 4.398,
|
| 1028 |
+
"eval_steps_per_second": 1.467,
|
| 1029 |
+
"step": 6800
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 0.013324897235753287,
|
| 1033 |
+
"grad_norm": 2.3113036155700684,
|
| 1034 |
+
"learning_rate": 6.6624181681246746e-06,
|
| 1035 |
+
"loss": 2.1643,
|
| 1036 |
+
"step": 6900
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 0.013324897235753287,
|
| 1040 |
+
"eval_loss": 2.2002882957458496,
|
| 1041 |
+
"eval_runtime": 454.153,
|
| 1042 |
+
"eval_samples_per_second": 4.408,
|
| 1043 |
+
"eval_steps_per_second": 1.471,
|
| 1044 |
+
"step": 6900
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 0.013518011688445363,
|
| 1048 |
+
"grad_norm": 2.2467916011810303,
|
| 1049 |
+
"learning_rate": 6.75897495316996e-06,
|
| 1050 |
+
"loss": 2.1609,
|
| 1051 |
+
"step": 7000
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 0.013518011688445363,
|
| 1055 |
+
"eval_loss": 2.1954784393310547,
|
| 1056 |
+
"eval_runtime": 455.4356,
|
| 1057 |
+
"eval_samples_per_second": 4.396,
|
| 1058 |
+
"eval_steps_per_second": 1.467,
|
| 1059 |
+
"step": 7000
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.01371112614113744,
|
| 1063 |
+
"grad_norm": 2.992947816848755,
|
| 1064 |
+
"learning_rate": 6.855531738215244e-06,
|
| 1065 |
+
"loss": 2.1617,
|
| 1066 |
+
"step": 7100
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.01371112614113744,
|
| 1070 |
+
"eval_loss": 2.192232608795166,
|
| 1071 |
+
"eval_runtime": 455.4904,
|
| 1072 |
+
"eval_samples_per_second": 4.395,
|
| 1073 |
+
"eval_steps_per_second": 1.467,
|
| 1074 |
+
"step": 7100
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.013904240593829517,
|
| 1078 |
+
"grad_norm": 2.2335407733917236,
|
| 1079 |
+
"learning_rate": 6.95208852326053e-06,
|
| 1080 |
+
"loss": 2.1525,
|
| 1081 |
+
"step": 7200
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.013904240593829517,
|
| 1085 |
+
"eval_loss": 2.1886117458343506,
|
| 1086 |
+
"eval_runtime": 451.074,
|
| 1087 |
+
"eval_samples_per_second": 4.438,
|
| 1088 |
+
"eval_steps_per_second": 1.481,
|
| 1089 |
+
"step": 7200
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 0.014097355046521593,
|
| 1093 |
+
"grad_norm": 2.345625877380371,
|
| 1094 |
+
"learning_rate": 7.0486453083058155e-06,
|
| 1095 |
+
"loss": 2.15,
|
| 1096 |
+
"step": 7300
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 0.014097355046521593,
|
| 1100 |
+
"eval_loss": 2.1845381259918213,
|
| 1101 |
+
"eval_runtime": 442.662,
|
| 1102 |
+
"eval_samples_per_second": 4.523,
|
| 1103 |
+
"eval_steps_per_second": 1.509,
|
| 1104 |
+
"step": 7300
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 0.01429046949921367,
|
| 1108 |
+
"grad_norm": 2.1838791370391846,
|
| 1109 |
+
"learning_rate": 7.1452020933511e-06,
|
| 1110 |
+
"loss": 2.1579,
|
| 1111 |
+
"step": 7400
|
| 1112 |
+
},
|
| 1113 |
+
{
|
| 1114 |
+
"epoch": 0.01429046949921367,
|
| 1115 |
+
"eval_loss": 2.1800084114074707,
|
| 1116 |
+
"eval_runtime": 448.4458,
|
| 1117 |
+
"eval_samples_per_second": 4.464,
|
| 1118 |
+
"eval_steps_per_second": 1.49,
|
| 1119 |
+
"step": 7400
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 0.014483583951905746,
|
| 1123 |
+
"grad_norm": 2.7497525215148926,
|
| 1124 |
+
"learning_rate": 7.241758878396384e-06,
|
| 1125 |
+
"loss": 2.1352,
|
| 1126 |
+
"step": 7500
|
| 1127 |
+
},
|
| 1128 |
+
{
|
| 1129 |
+
"epoch": 0.014483583951905746,
|
| 1130 |
+
"eval_loss": 2.178600311279297,
|
| 1131 |
+
"eval_runtime": 449.0785,
|
| 1132 |
+
"eval_samples_per_second": 4.458,
|
| 1133 |
+
"eval_steps_per_second": 1.487,
|
| 1134 |
+
"step": 7500
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 0.014676698404597824,
|
| 1138 |
+
"grad_norm": 5.904616355895996,
|
| 1139 |
+
"learning_rate": 7.338315663441671e-06,
|
| 1140 |
+
"loss": 2.151,
|
| 1141 |
+
"step": 7600
|
| 1142 |
+
},
|
| 1143 |
+
{
|
| 1144 |
+
"epoch": 0.014676698404597824,
|
| 1145 |
+
"eval_loss": 2.197683334350586,
|
| 1146 |
+
"eval_runtime": 449.0498,
|
| 1147 |
+
"eval_samples_per_second": 4.458,
|
| 1148 |
+
"eval_steps_per_second": 1.488,
|
| 1149 |
+
"step": 7600
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 0.0148698128572899,
|
| 1153 |
+
"grad_norm": 2.8403713703155518,
|
| 1154 |
+
"learning_rate": 7.434872448486956e-06,
|
| 1155 |
+
"loss": 2.1309,
|
| 1156 |
+
"step": 7700
|
| 1157 |
+
},
|
| 1158 |
+
{
|
| 1159 |
+
"epoch": 0.0148698128572899,
|
| 1160 |
+
"eval_loss": 2.1773977279663086,
|
| 1161 |
+
"eval_runtime": 451.4103,
|
| 1162 |
+
"eval_samples_per_second": 4.435,
|
| 1163 |
+
"eval_steps_per_second": 1.48,
|
| 1164 |
+
"step": 7700
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.015062927309981976,
|
| 1168 |
+
"grad_norm": 6.5780415534973145,
|
| 1169 |
+
"learning_rate": 7.53142923353224e-06,
|
| 1170 |
+
"loss": 2.1228,
|
| 1171 |
+
"step": 7800
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.015062927309981976,
|
| 1175 |
+
"eval_loss": 2.1752445697784424,
|
| 1176 |
+
"eval_runtime": 450.5777,
|
| 1177 |
+
"eval_samples_per_second": 4.443,
|
| 1178 |
+
"eval_steps_per_second": 1.483,
|
| 1179 |
+
"step": 7800
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.015256041762674054,
|
| 1183 |
+
"grad_norm": 1.9982277154922485,
|
| 1184 |
+
"learning_rate": 7.627986018577526e-06,
|
| 1185 |
+
"loss": 2.1401,
|
| 1186 |
+
"step": 7900
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.015256041762674054,
|
| 1190 |
+
"eval_loss": 2.170668363571167,
|
| 1191 |
+
"eval_runtime": 453.2726,
|
| 1192 |
+
"eval_samples_per_second": 4.417,
|
| 1193 |
+
"eval_steps_per_second": 1.474,
|
| 1194 |
+
"step": 7900
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 0.01544915621536613,
|
| 1198 |
+
"grad_norm": 2.037261486053467,
|
| 1199 |
+
"learning_rate": 7.72454280362281e-06,
|
| 1200 |
+
"loss": 2.1276,
|
| 1201 |
+
"step": 8000
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"epoch": 0.01544915621536613,
|
| 1205 |
+
"eval_loss": 2.166703462600708,
|
| 1206 |
+
"eval_runtime": 451.1428,
|
| 1207 |
+
"eval_samples_per_second": 4.438,
|
| 1208 |
+
"eval_steps_per_second": 1.481,
|
| 1209 |
+
"step": 8000
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 0.015642270668058206,
|
| 1213 |
+
"grad_norm": 2.6199100017547607,
|
| 1214 |
+
"learning_rate": 7.821099588668096e-06,
|
| 1215 |
+
"loss": 2.1366,
|
| 1216 |
+
"step": 8100
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 0.015642270668058206,
|
| 1220 |
+
"eval_loss": 2.1716530323028564,
|
| 1221 |
+
"eval_runtime": 450.9985,
|
| 1222 |
+
"eval_samples_per_second": 4.439,
|
| 1223 |
+
"eval_steps_per_second": 1.481,
|
| 1224 |
+
"step": 8100
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 0.01583538512075028,
|
| 1228 |
+
"grad_norm": 2.100360870361328,
|
| 1229 |
+
"learning_rate": 7.917656373713381e-06,
|
| 1230 |
+
"loss": 2.1219,
|
| 1231 |
+
"step": 8200
|
| 1232 |
+
},
|
| 1233 |
+
{
|
| 1234 |
+
"epoch": 0.01583538512075028,
|
| 1235 |
+
"eval_loss": 2.1657602787017822,
|
| 1236 |
+
"eval_runtime": 449.8619,
|
| 1237 |
+
"eval_samples_per_second": 4.45,
|
| 1238 |
+
"eval_steps_per_second": 1.485,
|
| 1239 |
+
"step": 8200
|
| 1240 |
+
},
|
| 1241 |
+
{
|
| 1242 |
+
"epoch": 0.01602849957344236,
|
| 1243 |
+
"grad_norm": 2.637589931488037,
|
| 1244 |
+
"learning_rate": 8.014213158758666e-06,
|
| 1245 |
+
"loss": 2.1232,
|
| 1246 |
+
"step": 8300
|
| 1247 |
+
},
|
| 1248 |
+
{
|
| 1249 |
+
"epoch": 0.01602849957344236,
|
| 1250 |
+
"eval_loss": 2.1619670391082764,
|
| 1251 |
+
"eval_runtime": 444.6116,
|
| 1252 |
+
"eval_samples_per_second": 4.503,
|
| 1253 |
+
"eval_steps_per_second": 1.502,
|
| 1254 |
+
"step": 8300
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 0.016221614026134437,
|
| 1258 |
+
"grad_norm": 2.937350273132324,
|
| 1259 |
+
"learning_rate": 8.110769943803951e-06,
|
| 1260 |
+
"loss": 2.1195,
|
| 1261 |
+
"step": 8400
|
| 1262 |
+
},
|
| 1263 |
+
{
|
| 1264 |
+
"epoch": 0.016221614026134437,
|
| 1265 |
+
"eval_loss": 2.160137891769409,
|
| 1266 |
+
"eval_runtime": 452.1301,
|
| 1267 |
+
"eval_samples_per_second": 4.428,
|
| 1268 |
+
"eval_steps_per_second": 1.477,
|
| 1269 |
+
"step": 8400
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.016414728478826513,
|
| 1273 |
+
"grad_norm": 2.5765292644500732,
|
| 1274 |
+
"learning_rate": 8.207326728849237e-06,
|
| 1275 |
+
"loss": 2.124,
|
| 1276 |
+
"step": 8500
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.016414728478826513,
|
| 1280 |
+
"eval_loss": 2.1571407318115234,
|
| 1281 |
+
"eval_runtime": 449.1301,
|
| 1282 |
+
"eval_samples_per_second": 4.458,
|
| 1283 |
+
"eval_steps_per_second": 1.487,
|
| 1284 |
+
"step": 8500
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.01660784293151859,
|
| 1288 |
+
"grad_norm": 3.157248020172119,
|
| 1289 |
+
"learning_rate": 8.303883513894522e-06,
|
| 1290 |
+
"loss": 2.1231,
|
| 1291 |
+
"step": 8600
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.01660784293151859,
|
| 1295 |
+
"eval_loss": 2.1576952934265137,
|
| 1296 |
+
"eval_runtime": 450.206,
|
| 1297 |
+
"eval_samples_per_second": 4.447,
|
| 1298 |
+
"eval_steps_per_second": 1.484,
|
| 1299 |
+
"step": 8600
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 0.016800957384210665,
|
| 1303 |
+
"grad_norm": 2.0448155403137207,
|
| 1304 |
+
"learning_rate": 8.400440298939807e-06,
|
| 1305 |
+
"loss": 2.1095,
|
| 1306 |
+
"step": 8700
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 0.016800957384210665,
|
| 1310 |
+
"eval_loss": 2.1506128311157227,
|
| 1311 |
+
"eval_runtime": 459.0304,
|
| 1312 |
+
"eval_samples_per_second": 4.361,
|
| 1313 |
+
"eval_steps_per_second": 1.455,
|
| 1314 |
+
"step": 8700
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 0.016994071836902745,
|
| 1318 |
+
"grad_norm": 2.436048746109009,
|
| 1319 |
+
"learning_rate": 8.496997083985092e-06,
|
| 1320 |
+
"loss": 2.1052,
|
| 1321 |
+
"step": 8800
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 0.016994071836902745,
|
| 1325 |
+
"eval_loss": 2.1523828506469727,
|
| 1326 |
+
"eval_runtime": 448.3924,
|
| 1327 |
+
"eval_samples_per_second": 4.465,
|
| 1328 |
+
"eval_steps_per_second": 1.49,
|
| 1329 |
+
"step": 8800
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 0.01718718628959482,
|
| 1333 |
+
"grad_norm": 2.020001173019409,
|
| 1334 |
+
"learning_rate": 8.593553869030378e-06,
|
| 1335 |
+
"loss": 2.1114,
|
| 1336 |
+
"step": 8900
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 0.01718718628959482,
|
| 1340 |
+
"eval_loss": 2.149531126022339,
|
| 1341 |
+
"eval_runtime": 473.4917,
|
| 1342 |
+
"eval_samples_per_second": 4.228,
|
| 1343 |
+
"eval_steps_per_second": 1.411,
|
| 1344 |
+
"step": 8900
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 0.017380300742286896,
|
| 1348 |
+
"grad_norm": 1.7989071607589722,
|
| 1349 |
+
"learning_rate": 8.690110654075663e-06,
|
| 1350 |
+
"loss": 2.1113,
|
| 1351 |
+
"step": 9000
|
| 1352 |
+
},
|
| 1353 |
+
{
|
| 1354 |
+
"epoch": 0.017380300742286896,
|
| 1355 |
+
"eval_loss": 2.145468235015869,
|
| 1356 |
+
"eval_runtime": 447.8226,
|
| 1357 |
+
"eval_samples_per_second": 4.471,
|
| 1358 |
+
"eval_steps_per_second": 1.492,
|
| 1359 |
+
"step": 9000
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 0.017573415194978972,
|
| 1363 |
+
"grad_norm": 1.8341110944747925,
|
| 1364 |
+
"learning_rate": 8.786667439120948e-06,
|
| 1365 |
+
"loss": 2.1072,
|
| 1366 |
+
"step": 9100
|
| 1367 |
+
},
|
| 1368 |
+
{
|
| 1369 |
+
"epoch": 0.017573415194978972,
|
| 1370 |
+
"eval_loss": 2.1460776329040527,
|
| 1371 |
+
"eval_runtime": 449.2918,
|
| 1372 |
+
"eval_samples_per_second": 4.456,
|
| 1373 |
+
"eval_steps_per_second": 1.487,
|
| 1374 |
+
"step": 9100
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.01776652964767105,
|
| 1378 |
+
"grad_norm": 2.2147250175476074,
|
| 1379 |
+
"learning_rate": 8.883224224166233e-06,
|
| 1380 |
+
"loss": 2.1023,
|
| 1381 |
+
"step": 9200
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.01776652964767105,
|
| 1385 |
+
"eval_loss": 2.144148349761963,
|
| 1386 |
+
"eval_runtime": 449.952,
|
| 1387 |
+
"eval_samples_per_second": 4.449,
|
| 1388 |
+
"eval_steps_per_second": 1.485,
|
| 1389 |
+
"step": 9200
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.017959644100363124,
|
| 1393 |
+
"grad_norm": 1.647185206413269,
|
| 1394 |
+
"learning_rate": 8.979781009211517e-06,
|
| 1395 |
+
"loss": 2.1091,
|
| 1396 |
+
"step": 9300
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.017959644100363124,
|
| 1400 |
+
"eval_loss": 2.1457300186157227,
|
| 1401 |
+
"eval_runtime": 453.9364,
|
| 1402 |
+
"eval_samples_per_second": 4.41,
|
| 1403 |
+
"eval_steps_per_second": 1.472,
|
| 1404 |
+
"step": 9300
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 0.018152758553055204,
|
| 1408 |
+
"grad_norm": 1.5562564134597778,
|
| 1409 |
+
"learning_rate": 9.076337794256804e-06,
|
| 1410 |
+
"loss": 2.097,
|
| 1411 |
+
"step": 9400
|
| 1412 |
+
},
|
| 1413 |
+
{
|
| 1414 |
+
"epoch": 0.018152758553055204,
|
| 1415 |
+
"eval_loss": 2.13842511177063,
|
| 1416 |
+
"eval_runtime": 451.2721,
|
| 1417 |
+
"eval_samples_per_second": 4.436,
|
| 1418 |
+
"eval_steps_per_second": 1.48,
|
| 1419 |
+
"step": 9400
|
| 1420 |
+
},
|
| 1421 |
+
{
|
| 1422 |
+
"epoch": 0.01834587300574728,
|
| 1423 |
+
"grad_norm": 1.7713780403137207,
|
| 1424 |
+
"learning_rate": 9.172894579302089e-06,
|
| 1425 |
+
"loss": 2.0898,
|
| 1426 |
+
"step": 9500
|
| 1427 |
+
},
|
| 1428 |
+
{
|
| 1429 |
+
"epoch": 0.01834587300574728,
|
| 1430 |
+
"eval_loss": 2.1376545429229736,
|
| 1431 |
+
"eval_runtime": 448.606,
|
| 1432 |
+
"eval_samples_per_second": 4.463,
|
| 1433 |
+
"eval_steps_per_second": 1.489,
|
| 1434 |
+
"step": 9500
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 0.018538987458439356,
|
| 1438 |
+
"grad_norm": 2.3007733821868896,
|
| 1439 |
+
"learning_rate": 9.269451364347373e-06,
|
| 1440 |
+
"loss": 2.093,
|
| 1441 |
+
"step": 9600
|
| 1442 |
+
},
|
| 1443 |
+
{
|
| 1444 |
+
"epoch": 0.018538987458439356,
|
| 1445 |
+
"eval_loss": 2.1340465545654297,
|
| 1446 |
+
"eval_runtime": 446.1625,
|
| 1447 |
+
"eval_samples_per_second": 4.487,
|
| 1448 |
+
"eval_steps_per_second": 1.497,
|
| 1449 |
+
"step": 9600
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 0.01873210191113143,
|
| 1453 |
+
"grad_norm": 1.4584991931915283,
|
| 1454 |
+
"learning_rate": 9.366008149392658e-06,
|
| 1455 |
+
"loss": 2.0982,
|
| 1456 |
+
"step": 9700
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 0.01873210191113143,
|
| 1460 |
+
"eval_loss": 2.1380152702331543,
|
| 1461 |
+
"eval_runtime": 448.4165,
|
| 1462 |
+
"eval_samples_per_second": 4.465,
|
| 1463 |
+
"eval_steps_per_second": 1.49,
|
| 1464 |
+
"step": 9700
|
| 1465 |
+
},
|
| 1466 |
+
{
|
| 1467 |
+
"epoch": 0.018925216363823508,
|
| 1468 |
+
"grad_norm": 1.7227883338928223,
|
| 1469 |
+
"learning_rate": 9.462564934437943e-06,
|
| 1470 |
+
"loss": 2.0887,
|
| 1471 |
+
"step": 9800
|
| 1472 |
+
},
|
| 1473 |
+
{
|
| 1474 |
+
"epoch": 0.018925216363823508,
|
| 1475 |
+
"eval_loss": 2.128800868988037,
|
| 1476 |
+
"eval_runtime": 448.7222,
|
| 1477 |
+
"eval_samples_per_second": 4.462,
|
| 1478 |
+
"eval_steps_per_second": 1.489,
|
| 1479 |
+
"step": 9800
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.019118330816515587,
|
| 1483 |
+
"grad_norm": 7.744117736816406,
|
| 1484 |
+
"learning_rate": 9.559121719483228e-06,
|
| 1485 |
+
"loss": 2.0938,
|
| 1486 |
+
"step": 9900
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.019118330816515587,
|
| 1490 |
+
"eval_loss": 2.1388182640075684,
|
| 1491 |
+
"eval_runtime": 449.3968,
|
| 1492 |
+
"eval_samples_per_second": 4.455,
|
| 1493 |
+
"eval_steps_per_second": 1.486,
|
| 1494 |
+
"step": 9900
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 0.019311445269207663,
|
| 1498 |
+
"grad_norm": 1.3925788402557373,
|
| 1499 |
+
"learning_rate": 9.655678504528514e-06,
|
| 1500 |
+
"loss": 2.0923,
|
| 1501 |
+
"step": 10000
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 0.019311445269207663,
|
| 1505 |
+
"eval_loss": 2.127917528152466,
|
| 1506 |
+
"eval_runtime": 474.9682,
|
| 1507 |
+
"eval_samples_per_second": 4.215,
|
| 1508 |
+
"eval_steps_per_second": 1.406,
|
| 1509 |
+
"step": 10000
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 0.01950455972189974,
|
| 1513 |
+
"grad_norm": 1.4508131742477417,
|
| 1514 |
+
"learning_rate": 9.752235289573799e-06,
|
| 1515 |
+
"loss": 2.1154,
|
| 1516 |
+
"step": 10100
|
| 1517 |
+
},
|
| 1518 |
+
{
|
| 1519 |
+
"epoch": 0.01950455972189974,
|
| 1520 |
+
"eval_loss": 2.132688045501709,
|
| 1521 |
+
"eval_runtime": 747.3867,
|
| 1522 |
+
"eval_samples_per_second": 2.679,
|
| 1523 |
+
"eval_steps_per_second": 0.894,
|
| 1524 |
+
"step": 10100
|
| 1525 |
+
},
|
| 1526 |
+
{
|
| 1527 |
+
"epoch": 0.019697674174591815,
|
| 1528 |
+
"grad_norm": 1.6229538917541504,
|
| 1529 |
+
"learning_rate": 9.848792074619084e-06,
|
| 1530 |
+
"loss": 2.0896,
|
| 1531 |
+
"step": 10200
|
| 1532 |
+
},
|
| 1533 |
+
{
|
| 1534 |
+
"epoch": 0.019697674174591815,
|
| 1535 |
+
"eval_loss": 2.1259098052978516,
|
| 1536 |
+
"eval_runtime": 746.7156,
|
| 1537 |
+
"eval_samples_per_second": 2.681,
|
| 1538 |
+
"eval_steps_per_second": 0.895,
|
| 1539 |
+
"step": 10200
|
| 1540 |
+
},
|
| 1541 |
+
{
|
| 1542 |
+
"epoch": 0.01989078862728389,
|
| 1543 |
+
"grad_norm": 1.4115490913391113,
|
| 1544 |
+
"learning_rate": 9.94534885966437e-06,
|
| 1545 |
+
"loss": 2.0825,
|
| 1546 |
+
"step": 10300
|
| 1547 |
+
},
|
| 1548 |
+
{
|
| 1549 |
+
"epoch": 0.01989078862728389,
|
| 1550 |
+
"eval_loss": 2.119873523712158,
|
| 1551 |
+
"eval_runtime": 746.4042,
|
| 1552 |
+
"eval_samples_per_second": 2.682,
|
| 1553 |
+
"eval_steps_per_second": 0.895,
|
| 1554 |
+
"step": 10300
|
| 1555 |
+
},
|
| 1556 |
+
{
|
| 1557 |
+
"epoch": 0.02008390307997597,
|
| 1558 |
+
"grad_norm": 1.349443793296814,
|
| 1559 |
+
"learning_rate": 1.0041905644709654e-05,
|
| 1560 |
+
"loss": 2.0791,
|
| 1561 |
+
"step": 10400
|
| 1562 |
+
},
|
| 1563 |
+
{
|
| 1564 |
+
"epoch": 0.02008390307997597,
|
| 1565 |
+
"eval_loss": 2.1162357330322266,
|
| 1566 |
+
"eval_runtime": 745.7602,
|
| 1567 |
+
"eval_samples_per_second": 2.685,
|
| 1568 |
+
"eval_steps_per_second": 0.896,
|
| 1569 |
+
"step": 10400
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"epoch": 0.020277017532668046,
|
| 1573 |
+
"grad_norm": 1.5818018913269043,
|
| 1574 |
+
"learning_rate": 1.013846242975494e-05,
|
| 1575 |
+
"loss": 2.0742,
|
| 1576 |
+
"step": 10500
|
| 1577 |
+
},
|
| 1578 |
+
{
|
| 1579 |
+
"epoch": 0.020277017532668046,
|
| 1580 |
+
"eval_loss": 2.1109495162963867,
|
| 1581 |
+
"eval_runtime": 751.1179,
|
| 1582 |
+
"eval_samples_per_second": 2.665,
|
| 1583 |
+
"eval_steps_per_second": 0.889,
|
| 1584 |
+
"step": 10500
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.020470131985360122,
|
| 1588 |
+
"grad_norm": 1.2264171838760376,
|
| 1589 |
+
"learning_rate": 1.0235019214800225e-05,
|
| 1590 |
+
"loss": 2.069,
|
| 1591 |
+
"step": 10600
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.020470131985360122,
|
| 1595 |
+
"eval_loss": 2.108383893966675,
|
| 1596 |
+
"eval_runtime": 745.7151,
|
| 1597 |
+
"eval_samples_per_second": 2.685,
|
| 1598 |
+
"eval_steps_per_second": 0.896,
|
| 1599 |
+
"step": 10600
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 0.0206632464380522,
|
| 1603 |
+
"grad_norm": 1.3637938499450684,
|
| 1604 |
+
"learning_rate": 1.033157599984551e-05,
|
| 1605 |
+
"loss": 2.0665,
|
| 1606 |
+
"step": 10700
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 0.0206632464380522,
|
| 1610 |
+
"eval_loss": 2.1054770946502686,
|
| 1611 |
+
"eval_runtime": 747.0672,
|
| 1612 |
+
"eval_samples_per_second": 2.68,
|
| 1613 |
+
"eval_steps_per_second": 0.894,
|
| 1614 |
+
"step": 10700
|
| 1615 |
+
},
|
| 1616 |
+
{
|
| 1617 |
+
"epoch": 0.020856360890744274,
|
| 1618 |
+
"grad_norm": 1.5810554027557373,
|
| 1619 |
+
"learning_rate": 1.0428132784890794e-05,
|
| 1620 |
+
"loss": 2.0599,
|
| 1621 |
+
"step": 10800
|
| 1622 |
+
},
|
| 1623 |
+
{
|
| 1624 |
+
"epoch": 0.020856360890744274,
|
| 1625 |
+
"eval_loss": 2.106874704360962,
|
| 1626 |
+
"eval_runtime": 749.8957,
|
| 1627 |
+
"eval_samples_per_second": 2.67,
|
| 1628 |
+
"eval_steps_per_second": 0.891,
|
| 1629 |
+
"step": 10800
|
| 1630 |
+
},
|
| 1631 |
+
{
|
| 1632 |
+
"epoch": 0.02104947534343635,
|
| 1633 |
+
"grad_norm": 1.1894757747650146,
|
| 1634 |
+
"learning_rate": 1.052468956993608e-05,
|
| 1635 |
+
"loss": 2.0616,
|
| 1636 |
+
"step": 10900
|
| 1637 |
+
},
|
| 1638 |
+
{
|
| 1639 |
+
"epoch": 0.02104947534343635,
|
| 1640 |
+
"eval_loss": 2.0996124744415283,
|
| 1641 |
+
"eval_runtime": 747.4095,
|
| 1642 |
+
"eval_samples_per_second": 2.679,
|
| 1643 |
+
"eval_steps_per_second": 0.894,
|
| 1644 |
+
"step": 10900
|
| 1645 |
+
},
|
| 1646 |
+
{
|
| 1647 |
+
"epoch": 0.02124258979612843,
|
| 1648 |
+
"grad_norm": 1.0796828269958496,
|
| 1649 |
+
"learning_rate": 1.0621246354981366e-05,
|
| 1650 |
+
"loss": 2.0657,
|
| 1651 |
+
"step": 11000
|
| 1652 |
+
},
|
| 1653 |
+
{
|
| 1654 |
+
"epoch": 0.02124258979612843,
|
| 1655 |
+
"eval_loss": 2.098177671432495,
|
| 1656 |
+
"eval_runtime": 748.3753,
|
| 1657 |
+
"eval_samples_per_second": 2.675,
|
| 1658 |
+
"eval_steps_per_second": 0.893,
|
| 1659 |
+
"step": 11000
|
| 1660 |
+
},
|
| 1661 |
+
{
|
| 1662 |
+
"epoch": 0.021435704248820506,
|
| 1663 |
+
"grad_norm": 1.5541911125183105,
|
| 1664 |
+
"learning_rate": 1.071780314002665e-05,
|
| 1665 |
+
"loss": 2.0451,
|
| 1666 |
+
"step": 11100
|
| 1667 |
+
},
|
| 1668 |
+
{
|
| 1669 |
+
"epoch": 0.021435704248820506,
|
| 1670 |
+
"eval_loss": 2.0963504314422607,
|
| 1671 |
+
"eval_runtime": 445.3814,
|
| 1672 |
+
"eval_samples_per_second": 4.495,
|
| 1673 |
+
"eval_steps_per_second": 1.5,
|
| 1674 |
+
"step": 11100
|
| 1675 |
+
},
|
| 1676 |
+
{
|
| 1677 |
+
"epoch": 0.02162881870151258,
|
| 1678 |
+
"grad_norm": 1.224878191947937,
|
| 1679 |
+
"learning_rate": 1.0814359925071936e-05,
|
| 1680 |
+
"loss": 2.0636,
|
| 1681 |
+
"step": 11200
|
| 1682 |
+
},
|
| 1683 |
+
{
|
| 1684 |
+
"epoch": 0.02162881870151258,
|
| 1685 |
+
"eval_loss": 2.0935559272766113,
|
| 1686 |
+
"eval_runtime": 446.9887,
|
| 1687 |
+
"eval_samples_per_second": 4.479,
|
| 1688 |
+
"eval_steps_per_second": 1.494,
|
| 1689 |
+
"step": 11200
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.021821933154204658,
|
| 1693 |
+
"grad_norm": 1.1165440082550049,
|
| 1694 |
+
"learning_rate": 1.0910916710117222e-05,
|
| 1695 |
+
"loss": 2.0675,
|
| 1696 |
+
"step": 11300
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.021821933154204658,
|
| 1700 |
+
"eval_loss": 2.096463203430176,
|
| 1701 |
+
"eval_runtime": 442.2835,
|
| 1702 |
+
"eval_samples_per_second": 4.527,
|
| 1703 |
+
"eval_steps_per_second": 1.51,
|
| 1704 |
+
"step": 11300
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 0.022015047606896734,
|
| 1708 |
+
"grad_norm": 1.2041581869125366,
|
| 1709 |
+
"learning_rate": 1.1007473495162505e-05,
|
| 1710 |
+
"loss": 2.0488,
|
| 1711 |
+
"step": 11400
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 0.022015047606896734,
|
| 1715 |
+
"eval_loss": 2.092092990875244,
|
| 1716 |
+
"eval_runtime": 441.9553,
|
| 1717 |
+
"eval_samples_per_second": 4.53,
|
| 1718 |
+
"eval_steps_per_second": 1.511,
|
| 1719 |
+
"step": 11400
|
| 1720 |
+
},
|
| 1721 |
+
{
|
| 1722 |
+
"epoch": 0.022208162059588813,
|
| 1723 |
+
"grad_norm": 1.3523286581039429,
|
| 1724 |
+
"learning_rate": 1.110403028020779e-05,
|
| 1725 |
+
"loss": 2.05,
|
| 1726 |
+
"step": 11500
|
| 1727 |
+
},
|
| 1728 |
+
{
|
| 1729 |
+
"epoch": 0.022208162059588813,
|
| 1730 |
+
"eval_loss": 2.092932939529419,
|
| 1731 |
+
"eval_runtime": 445.7387,
|
| 1732 |
+
"eval_samples_per_second": 4.491,
|
| 1733 |
+
"eval_steps_per_second": 1.499,
|
| 1734 |
+
"step": 11500
|
| 1735 |
+
},
|
| 1736 |
+
{
|
| 1737 |
+
"epoch": 0.02240127651228089,
|
| 1738 |
+
"grad_norm": 1.2852915525436401,
|
| 1739 |
+
"learning_rate": 1.1200587065253076e-05,
|
| 1740 |
+
"loss": 2.0451,
|
| 1741 |
+
"step": 11600
|
| 1742 |
+
},
|
| 1743 |
+
{
|
| 1744 |
+
"epoch": 0.02240127651228089,
|
| 1745 |
+
"eval_loss": 2.08709716796875,
|
| 1746 |
+
"eval_runtime": 442.9466,
|
| 1747 |
+
"eval_samples_per_second": 4.52,
|
| 1748 |
+
"eval_steps_per_second": 1.508,
|
| 1749 |
+
"step": 11600
|
| 1750 |
+
},
|
| 1751 |
+
{
|
| 1752 |
+
"epoch": 0.022594390964972965,
|
| 1753 |
+
"grad_norm": 1.2034499645233154,
|
| 1754 |
+
"learning_rate": 1.129714385029836e-05,
|
| 1755 |
+
"loss": 2.05,
|
| 1756 |
+
"step": 11700
|
| 1757 |
+
},
|
| 1758 |
+
{
|
| 1759 |
+
"epoch": 0.022594390964972965,
|
| 1760 |
+
"eval_loss": 2.087824821472168,
|
| 1761 |
+
"eval_runtime": 441.7707,
|
| 1762 |
+
"eval_samples_per_second": 4.532,
|
| 1763 |
+
"eval_steps_per_second": 1.512,
|
| 1764 |
+
"step": 11700
|
| 1765 |
+
},
|
| 1766 |
+
{
|
| 1767 |
+
"epoch": 0.02278750541766504,
|
| 1768 |
+
"grad_norm": 4.378060340881348,
|
| 1769 |
+
"learning_rate": 1.1393700635343646e-05,
|
| 1770 |
+
"loss": 2.0418,
|
| 1771 |
+
"step": 11800
|
| 1772 |
+
},
|
| 1773 |
+
{
|
| 1774 |
+
"epoch": 0.02278750541766504,
|
| 1775 |
+
"eval_loss": 2.086357593536377,
|
| 1776 |
+
"eval_runtime": 443.3936,
|
| 1777 |
+
"eval_samples_per_second": 4.515,
|
| 1778 |
+
"eval_steps_per_second": 1.507,
|
| 1779 |
+
"step": 11800
|
| 1780 |
+
},
|
| 1781 |
+
{
|
| 1782 |
+
"epoch": 0.022980619870357117,
|
| 1783 |
+
"grad_norm": 1.2029449939727783,
|
| 1784 |
+
"learning_rate": 1.1490257420388931e-05,
|
| 1785 |
+
"loss": 2.0516,
|
| 1786 |
+
"step": 11900
|
| 1787 |
+
},
|
| 1788 |
+
{
|
| 1789 |
+
"epoch": 0.022980619870357117,
|
| 1790 |
+
"eval_loss": 2.084092140197754,
|
| 1791 |
+
"eval_runtime": 443.8581,
|
| 1792 |
+
"eval_samples_per_second": 4.51,
|
| 1793 |
+
"eval_steps_per_second": 1.505,
|
| 1794 |
+
"step": 11900
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.023173734323049196,
|
| 1798 |
+
"grad_norm": 1.160551905632019,
|
| 1799 |
+
"learning_rate": 1.1586814205434217e-05,
|
| 1800 |
+
"loss": 2.0531,
|
| 1801 |
+
"step": 12000
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.023173734323049196,
|
| 1805 |
+
"eval_loss": 2.0821826457977295,
|
| 1806 |
+
"eval_runtime": 450.152,
|
| 1807 |
+
"eval_samples_per_second": 4.447,
|
| 1808 |
+
"eval_steps_per_second": 1.484,
|
| 1809 |
+
"step": 12000
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 0.023366848775741272,
|
| 1813 |
+
"grad_norm": 1.156447410583496,
|
| 1814 |
+
"learning_rate": 1.1683370990479502e-05,
|
| 1815 |
+
"loss": 2.0445,
|
| 1816 |
+
"step": 12100
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 0.023366848775741272,
|
| 1820 |
+
"eval_loss": 2.079378366470337,
|
| 1821 |
+
"eval_runtime": 442.5582,
|
| 1822 |
+
"eval_samples_per_second": 4.524,
|
| 1823 |
+
"eval_steps_per_second": 1.509,
|
| 1824 |
+
"step": 12100
|
| 1825 |
+
},
|
| 1826 |
+
{
|
| 1827 |
+
"epoch": 0.02355996322843335,
|
| 1828 |
+
"grad_norm": 1.1055281162261963,
|
| 1829 |
+
"learning_rate": 1.1779927775524787e-05,
|
| 1830 |
+
"loss": 2.0482,
|
| 1831 |
+
"step": 12200
|
| 1832 |
+
},
|
| 1833 |
+
{
|
| 1834 |
+
"epoch": 0.02355996322843335,
|
| 1835 |
+
"eval_loss": 2.075855016708374,
|
| 1836 |
+
"eval_runtime": 442.7884,
|
| 1837 |
+
"eval_samples_per_second": 4.521,
|
| 1838 |
+
"eval_steps_per_second": 1.509,
|
| 1839 |
+
"step": 12200
|
| 1840 |
+
},
|
| 1841 |
+
{
|
| 1842 |
+
"epoch": 0.023753077681125424,
|
| 1843 |
+
"grad_norm": 1.0532505512237549,
|
| 1844 |
+
"learning_rate": 1.187648456057007e-05,
|
| 1845 |
+
"loss": 2.0446,
|
| 1846 |
+
"step": 12300
|
| 1847 |
+
},
|
| 1848 |
+
{
|
| 1849 |
+
"epoch": 0.023753077681125424,
|
| 1850 |
+
"eval_loss": 2.0749075412750244,
|
| 1851 |
+
"eval_runtime": 443.7026,
|
| 1852 |
+
"eval_samples_per_second": 4.512,
|
| 1853 |
+
"eval_steps_per_second": 1.506,
|
| 1854 |
+
"step": 12300
|
| 1855 |
+
},
|
| 1856 |
+
{
|
| 1857 |
+
"epoch": 0.0239461921338175,
|
| 1858 |
+
"grad_norm": 0.9835991859436035,
|
| 1859 |
+
"learning_rate": 1.1973041345615357e-05,
|
| 1860 |
+
"loss": 2.0393,
|
| 1861 |
+
"step": 12400
|
| 1862 |
+
},
|
| 1863 |
+
{
|
| 1864 |
+
"epoch": 0.0239461921338175,
|
| 1865 |
+
"eval_loss": 2.072516918182373,
|
| 1866 |
+
"eval_runtime": 442.5377,
|
| 1867 |
+
"eval_samples_per_second": 4.524,
|
| 1868 |
+
"eval_steps_per_second": 1.509,
|
| 1869 |
+
"step": 12400
|
| 1870 |
+
},
|
| 1871 |
+
{
|
| 1872 |
+
"epoch": 0.024139306586509576,
|
| 1873 |
+
"grad_norm": 1.1653251647949219,
|
| 1874 |
+
"learning_rate": 1.2069598130660643e-05,
|
| 1875 |
+
"loss": 2.0376,
|
| 1876 |
+
"step": 12500
|
| 1877 |
+
},
|
| 1878 |
+
{
|
| 1879 |
+
"epoch": 0.024139306586509576,
|
| 1880 |
+
"eval_loss": 2.0766026973724365,
|
| 1881 |
+
"eval_runtime": 442.5122,
|
| 1882 |
+
"eval_samples_per_second": 4.524,
|
| 1883 |
+
"eval_steps_per_second": 1.51,
|
| 1884 |
+
"step": 12500
|
| 1885 |
+
},
|
| 1886 |
+
{
|
| 1887 |
+
"epoch": 0.024332421039201656,
|
| 1888 |
+
"grad_norm": 1.0789021253585815,
|
| 1889 |
+
"learning_rate": 1.2166154915705926e-05,
|
| 1890 |
+
"loss": 2.0355,
|
| 1891 |
+
"step": 12600
|
| 1892 |
+
},
|
| 1893 |
+
{
|
| 1894 |
+
"epoch": 0.024332421039201656,
|
| 1895 |
+
"eval_loss": 2.0720646381378174,
|
| 1896 |
+
"eval_runtime": 440.799,
|
| 1897 |
+
"eval_samples_per_second": 4.542,
|
| 1898 |
+
"eval_steps_per_second": 1.515,
|
| 1899 |
+
"step": 12600
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.02452553549189373,
|
| 1903 |
+
"grad_norm": 1.171810507774353,
|
| 1904 |
+
"learning_rate": 1.2262711700751213e-05,
|
| 1905 |
+
"loss": 2.0397,
|
| 1906 |
+
"step": 12700
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.02452553549189373,
|
| 1910 |
+
"eval_loss": 2.0727598667144775,
|
| 1911 |
+
"eval_runtime": 442.7395,
|
| 1912 |
+
"eval_samples_per_second": 4.522,
|
| 1913 |
+
"eval_steps_per_second": 1.509,
|
| 1914 |
+
"step": 12700
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 0.024718649944585808,
|
| 1918 |
+
"grad_norm": 1.255620002746582,
|
| 1919 |
+
"learning_rate": 1.2359268485796498e-05,
|
| 1920 |
+
"loss": 2.0305,
|
| 1921 |
+
"step": 12800
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 0.024718649944585808,
|
| 1925 |
+
"eval_loss": 2.0683271884918213,
|
| 1926 |
+
"eval_runtime": 444.9818,
|
| 1927 |
+
"eval_samples_per_second": 4.499,
|
| 1928 |
+
"eval_steps_per_second": 1.501,
|
| 1929 |
+
"step": 12800
|
| 1930 |
+
},
|
| 1931 |
+
{
|
| 1932 |
+
"epoch": 0.024911764397277884,
|
| 1933 |
+
"grad_norm": 1.0595488548278809,
|
| 1934 |
+
"learning_rate": 1.2455825270841782e-05,
|
| 1935 |
+
"loss": 2.0362,
|
| 1936 |
+
"step": 12900
|
| 1937 |
+
},
|
| 1938 |
+
{
|
| 1939 |
+
"epoch": 0.024911764397277884,
|
| 1940 |
+
"eval_loss": 2.0663349628448486,
|
| 1941 |
+
"eval_runtime": 496.75,
|
| 1942 |
+
"eval_samples_per_second": 4.03,
|
| 1943 |
+
"eval_steps_per_second": 1.345,
|
| 1944 |
+
"step": 12900
|
| 1945 |
+
},
|
| 1946 |
+
{
|
| 1947 |
+
"epoch": 0.02510487884996996,
|
| 1948 |
+
"grad_norm": 1.1099809408187866,
|
| 1949 |
+
"learning_rate": 1.2552382055887069e-05,
|
| 1950 |
+
"loss": 2.0466,
|
| 1951 |
+
"step": 13000
|
| 1952 |
+
},
|
| 1953 |
+
{
|
| 1954 |
+
"epoch": 0.02510487884996996,
|
| 1955 |
+
"eval_loss": 2.064277410507202,
|
| 1956 |
+
"eval_runtime": 445.7251,
|
| 1957 |
+
"eval_samples_per_second": 4.492,
|
| 1958 |
+
"eval_steps_per_second": 1.499,
|
| 1959 |
+
"step": 13000
|
| 1960 |
+
},
|
| 1961 |
+
{
|
| 1962 |
+
"epoch": 0.02529799330266204,
|
| 1963 |
+
"grad_norm": 1.263329029083252,
|
| 1964 |
+
"learning_rate": 1.2648938840932354e-05,
|
| 1965 |
+
"loss": 2.0265,
|
| 1966 |
+
"step": 13100
|
| 1967 |
+
},
|
| 1968 |
+
{
|
| 1969 |
+
"epoch": 0.02529799330266204,
|
| 1970 |
+
"eval_loss": 2.062872886657715,
|
| 1971 |
+
"eval_runtime": 446.6164,
|
| 1972 |
+
"eval_samples_per_second": 4.483,
|
| 1973 |
+
"eval_steps_per_second": 1.496,
|
| 1974 |
+
"step": 13100
|
| 1975 |
+
},
|
| 1976 |
+
{
|
| 1977 |
+
"epoch": 0.025491107755354115,
|
| 1978 |
+
"grad_norm": 1.2612566947937012,
|
| 1979 |
+
"learning_rate": 1.2745495625977638e-05,
|
| 1980 |
+
"loss": 2.0343,
|
| 1981 |
+
"step": 13200
|
| 1982 |
+
},
|
| 1983 |
+
{
|
| 1984 |
+
"epoch": 0.025491107755354115,
|
| 1985 |
+
"eval_loss": 2.061845541000366,
|
| 1986 |
+
"eval_runtime": 444.0257,
|
| 1987 |
+
"eval_samples_per_second": 4.509,
|
| 1988 |
+
"eval_steps_per_second": 1.504,
|
| 1989 |
+
"step": 13200
|
| 1990 |
+
},
|
| 1991 |
+
{
|
| 1992 |
+
"epoch": 0.02568422220804619,
|
| 1993 |
+
"grad_norm": 1.1635104417800903,
|
| 1994 |
+
"learning_rate": 1.2842052411022923e-05,
|
| 1995 |
+
"loss": 2.0218,
|
| 1996 |
+
"step": 13300
|
| 1997 |
+
},
|
| 1998 |
+
{
|
| 1999 |
+
"epoch": 0.02568422220804619,
|
| 2000 |
+
"eval_loss": 2.0589962005615234,
|
| 2001 |
+
"eval_runtime": 447.3235,
|
| 2002 |
+
"eval_samples_per_second": 4.476,
|
| 2003 |
+
"eval_steps_per_second": 1.493,
|
| 2004 |
+
"step": 13300
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.025877336660738267,
|
| 2008 |
+
"grad_norm": 1.0829100608825684,
|
| 2009 |
+
"learning_rate": 1.2938609196068208e-05,
|
| 2010 |
+
"loss": 2.022,
|
| 2011 |
+
"step": 13400
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.025877336660738267,
|
| 2015 |
+
"eval_loss": 2.058737277984619,
|
| 2016 |
+
"eval_runtime": 441.9917,
|
| 2017 |
+
"eval_samples_per_second": 4.529,
|
| 2018 |
+
"eval_steps_per_second": 1.511,
|
| 2019 |
+
"step": 13400
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 0.026070451113430343,
|
| 2023 |
+
"grad_norm": 0.9413105249404907,
|
| 2024 |
+
"learning_rate": 1.3035165981113492e-05,
|
| 2025 |
+
"loss": 2.0146,
|
| 2026 |
+
"step": 13500
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 0.026070451113430343,
|
| 2030 |
+
"eval_loss": 2.0579559803009033,
|
| 2031 |
+
"eval_runtime": 454.9936,
|
| 2032 |
+
"eval_samples_per_second": 4.4,
|
| 2033 |
+
"eval_steps_per_second": 1.468,
|
| 2034 |
+
"step": 13500
|
| 2035 |
+
},
|
| 2036 |
+
{
|
| 2037 |
+
"epoch": 0.026263565566122422,
|
| 2038 |
+
"grad_norm": 0.9331383109092712,
|
| 2039 |
+
"learning_rate": 1.313172276615878e-05,
|
| 2040 |
+
"loss": 2.0266,
|
| 2041 |
+
"step": 13600
|
| 2042 |
+
},
|
| 2043 |
+
{
|
| 2044 |
+
"epoch": 0.026263565566122422,
|
| 2045 |
+
"eval_loss": 2.0555694103240967,
|
| 2046 |
+
"eval_runtime": 449.5577,
|
| 2047 |
+
"eval_samples_per_second": 4.453,
|
| 2048 |
+
"eval_steps_per_second": 1.486,
|
| 2049 |
+
"step": 13600
|
| 2050 |
+
},
|
| 2051 |
+
{
|
| 2052 |
+
"epoch": 0.0264566800188145,
|
| 2053 |
+
"grad_norm": 1.066735029220581,
|
| 2054 |
+
"learning_rate": 1.3228279551204064e-05,
|
| 2055 |
+
"loss": 2.0291,
|
| 2056 |
+
"step": 13700
|
| 2057 |
+
},
|
| 2058 |
+
{
|
| 2059 |
+
"epoch": 0.0264566800188145,
|
| 2060 |
+
"eval_loss": 2.054654598236084,
|
| 2061 |
+
"eval_runtime": 448.1523,
|
| 2062 |
+
"eval_samples_per_second": 4.467,
|
| 2063 |
+
"eval_steps_per_second": 1.491,
|
| 2064 |
+
"step": 13700
|
| 2065 |
+
},
|
| 2066 |
+
{
|
| 2067 |
+
"epoch": 0.026649794471506574,
|
| 2068 |
+
"grad_norm": 0.9554963111877441,
|
| 2069 |
+
"learning_rate": 1.3324836336249349e-05,
|
| 2070 |
+
"loss": 2.0108,
|
| 2071 |
+
"step": 13800
|
| 2072 |
+
},
|
| 2073 |
+
{
|
| 2074 |
+
"epoch": 0.026649794471506574,
|
| 2075 |
+
"eval_loss": 2.0530266761779785,
|
| 2076 |
+
"eval_runtime": 447.7752,
|
| 2077 |
+
"eval_samples_per_second": 4.471,
|
| 2078 |
+
"eval_steps_per_second": 1.492,
|
| 2079 |
+
"step": 13800
|
| 2080 |
+
},
|
| 2081 |
+
{
|
| 2082 |
+
"epoch": 0.02684290892419865,
|
| 2083 |
+
"grad_norm": 0.9701889753341675,
|
| 2084 |
+
"learning_rate": 1.3421393121294634e-05,
|
| 2085 |
+
"loss": 2.0306,
|
| 2086 |
+
"step": 13900
|
| 2087 |
+
},
|
| 2088 |
+
{
|
| 2089 |
+
"epoch": 0.02684290892419865,
|
| 2090 |
+
"eval_loss": 2.0509798526763916,
|
| 2091 |
+
"eval_runtime": 448.7204,
|
| 2092 |
+
"eval_samples_per_second": 4.462,
|
| 2093 |
+
"eval_steps_per_second": 1.489,
|
| 2094 |
+
"step": 13900
|
| 2095 |
+
},
|
| 2096 |
+
{
|
| 2097 |
+
"epoch": 0.027036023376890726,
|
| 2098 |
+
"grad_norm": 0.9781773090362549,
|
| 2099 |
+
"learning_rate": 1.351794990633992e-05,
|
| 2100 |
+
"loss": 2.0182,
|
| 2101 |
+
"step": 14000
|
| 2102 |
+
},
|
| 2103 |
+
{
|
| 2104 |
+
"epoch": 0.027036023376890726,
|
| 2105 |
+
"eval_loss": 2.0510973930358887,
|
| 2106 |
+
"eval_runtime": 449.0716,
|
| 2107 |
+
"eval_samples_per_second": 4.458,
|
| 2108 |
+
"eval_steps_per_second": 1.488,
|
| 2109 |
+
"step": 14000
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.027229137829582802,
|
| 2113 |
+
"grad_norm": 0.9371834397315979,
|
| 2114 |
+
"learning_rate": 1.3614506691385203e-05,
|
| 2115 |
+
"loss": 2.0255,
|
| 2116 |
+
"step": 14100
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.027229137829582802,
|
| 2120 |
+
"eval_loss": 2.048981189727783,
|
| 2121 |
+
"eval_runtime": 448.6448,
|
| 2122 |
+
"eval_samples_per_second": 4.462,
|
| 2123 |
+
"eval_steps_per_second": 1.489,
|
| 2124 |
+
"step": 14100
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 0.02742225228227488,
|
| 2128 |
+
"grad_norm": 0.8181398510932922,
|
| 2129 |
+
"learning_rate": 1.3711063476430488e-05,
|
| 2130 |
+
"loss": 2.0152,
|
| 2131 |
+
"step": 14200
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 0.02742225228227488,
|
| 2135 |
+
"eval_loss": 2.047391176223755,
|
| 2136 |
+
"eval_runtime": 442.5828,
|
| 2137 |
+
"eval_samples_per_second": 4.523,
|
| 2138 |
+
"eval_steps_per_second": 1.509,
|
| 2139 |
+
"step": 14200
|
| 2140 |
+
},
|
| 2141 |
+
{
|
| 2142 |
+
"epoch": 0.027615366734966958,
|
| 2143 |
+
"grad_norm": 0.9860553741455078,
|
| 2144 |
+
"learning_rate": 1.3807620261475774e-05,
|
| 2145 |
+
"loss": 2.0272,
|
| 2146 |
+
"step": 14300
|
| 2147 |
+
},
|
| 2148 |
+
{
|
| 2149 |
+
"epoch": 0.027615366734966958,
|
| 2150 |
+
"eval_loss": 2.047621250152588,
|
| 2151 |
+
"eval_runtime": 447.4945,
|
| 2152 |
+
"eval_samples_per_second": 4.474,
|
| 2153 |
+
"eval_steps_per_second": 1.493,
|
| 2154 |
+
"step": 14300
|
| 2155 |
+
},
|
| 2156 |
+
{
|
| 2157 |
+
"epoch": 0.027808481187659034,
|
| 2158 |
+
"grad_norm": 0.8708497881889343,
|
| 2159 |
+
"learning_rate": 1.390417704652106e-05,
|
| 2160 |
+
"loss": 2.0148,
|
| 2161 |
+
"step": 14400
|
| 2162 |
+
},
|
| 2163 |
+
{
|
| 2164 |
+
"epoch": 0.027808481187659034,
|
| 2165 |
+
"eval_loss": 2.0451223850250244,
|
| 2166 |
+
"eval_runtime": 442.4977,
|
| 2167 |
+
"eval_samples_per_second": 4.524,
|
| 2168 |
+
"eval_steps_per_second": 1.51,
|
| 2169 |
+
"step": 14400
|
| 2170 |
+
},
|
| 2171 |
+
{
|
| 2172 |
+
"epoch": 0.02800159564035111,
|
| 2173 |
+
"grad_norm": 0.7860450744628906,
|
| 2174 |
+
"learning_rate": 1.4000733831566346e-05,
|
| 2175 |
+
"loss": 2.0203,
|
| 2176 |
+
"step": 14500
|
| 2177 |
+
},
|
| 2178 |
+
{
|
| 2179 |
+
"epoch": 0.02800159564035111,
|
| 2180 |
+
"eval_loss": 2.0454912185668945,
|
| 2181 |
+
"eval_runtime": 446.3855,
|
| 2182 |
+
"eval_samples_per_second": 4.485,
|
| 2183 |
+
"eval_steps_per_second": 1.496,
|
| 2184 |
+
"step": 14500
|
| 2185 |
+
},
|
| 2186 |
+
{
|
| 2187 |
+
"epoch": 0.028194710093043186,
|
| 2188 |
+
"grad_norm": 0.9461175203323364,
|
| 2189 |
+
"learning_rate": 1.4097290616611631e-05,
|
| 2190 |
+
"loss": 2.0018,
|
| 2191 |
+
"step": 14600
|
| 2192 |
+
},
|
| 2193 |
+
{
|
| 2194 |
+
"epoch": 0.028194710093043186,
|
| 2195 |
+
"eval_loss": 2.0433313846588135,
|
| 2196 |
+
"eval_runtime": 444.3827,
|
| 2197 |
+
"eval_samples_per_second": 4.505,
|
| 2198 |
+
"eval_steps_per_second": 1.503,
|
| 2199 |
+
"step": 14600
|
| 2200 |
+
},
|
| 2201 |
+
{
|
| 2202 |
+
"epoch": 0.028387824545735265,
|
| 2203 |
+
"grad_norm": 0.9540334939956665,
|
| 2204 |
+
"learning_rate": 1.4193847401656915e-05,
|
| 2205 |
+
"loss": 2.0076,
|
| 2206 |
+
"step": 14700
|
| 2207 |
+
},
|
| 2208 |
+
{
|
| 2209 |
+
"epoch": 0.028387824545735265,
|
| 2210 |
+
"eval_loss": 2.0425238609313965,
|
| 2211 |
+
"eval_runtime": 446.3018,
|
| 2212 |
+
"eval_samples_per_second": 4.486,
|
| 2213 |
+
"eval_steps_per_second": 1.497,
|
| 2214 |
+
"step": 14700
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.02858093899842734,
|
| 2218 |
+
"grad_norm": 0.8437981605529785,
|
| 2219 |
+
"learning_rate": 1.42904041867022e-05,
|
| 2220 |
+
"loss": 2.0098,
|
| 2221 |
+
"step": 14800
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.02858093899842734,
|
| 2225 |
+
"eval_loss": 2.0422203540802,
|
| 2226 |
+
"eval_runtime": 442.7055,
|
| 2227 |
+
"eval_samples_per_second": 4.522,
|
| 2228 |
+
"eval_steps_per_second": 1.509,
|
| 2229 |
+
"step": 14800
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 0.028774053451119417,
|
| 2233 |
+
"grad_norm": 1.0205272436141968,
|
| 2234 |
+
"learning_rate": 1.4386960971747485e-05,
|
| 2235 |
+
"loss": 2.0116,
|
| 2236 |
+
"step": 14900
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 0.028774053451119417,
|
| 2240 |
+
"eval_loss": 2.040599822998047,
|
| 2241 |
+
"eval_runtime": 444.0043,
|
| 2242 |
+
"eval_samples_per_second": 4.509,
|
| 2243 |
+
"eval_steps_per_second": 1.504,
|
| 2244 |
+
"step": 14900
|
| 2245 |
+
},
|
| 2246 |
+
{
|
| 2247 |
+
"epoch": 0.028967167903811493,
|
| 2248 |
+
"grad_norm": 0.9906435608863831,
|
| 2249 |
+
"learning_rate": 1.4483517756792769e-05,
|
| 2250 |
+
"loss": 2.0212,
|
| 2251 |
+
"step": 15000
|
| 2252 |
+
},
|
| 2253 |
+
{
|
| 2254 |
+
"epoch": 0.028967167903811493,
|
| 2255 |
+
"eval_loss": 2.0395123958587646,
|
| 2256 |
+
"eval_runtime": 448.0337,
|
| 2257 |
+
"eval_samples_per_second": 4.468,
|
| 2258 |
+
"eval_steps_per_second": 1.491,
|
| 2259 |
+
"step": 15000
|
| 2260 |
+
},
|
| 2261 |
+
{
|
| 2262 |
+
"epoch": 0.02916028235650357,
|
| 2263 |
+
"grad_norm": 1.0708587169647217,
|
| 2264 |
+
"learning_rate": 1.4580074541838057e-05,
|
| 2265 |
+
"loss": 2.0119,
|
| 2266 |
+
"step": 15100
|
| 2267 |
+
},
|
| 2268 |
+
{
|
| 2269 |
+
"epoch": 0.02916028235650357,
|
| 2270 |
+
"eval_loss": 2.047750473022461,
|
| 2271 |
+
"eval_runtime": 443.2078,
|
| 2272 |
+
"eval_samples_per_second": 4.517,
|
| 2273 |
+
"eval_steps_per_second": 1.507,
|
| 2274 |
+
"step": 15100
|
| 2275 |
+
},
|
| 2276 |
+
{
|
| 2277 |
+
"epoch": 0.02935339680919565,
|
| 2278 |
+
"grad_norm": 1.0378001928329468,
|
| 2279 |
+
"learning_rate": 1.4676631326883342e-05,
|
| 2280 |
+
"loss": 2.0154,
|
| 2281 |
+
"step": 15200
|
| 2282 |
+
},
|
| 2283 |
+
{
|
| 2284 |
+
"epoch": 0.02935339680919565,
|
| 2285 |
+
"eval_loss": 2.0439839363098145,
|
| 2286 |
+
"eval_runtime": 441.3501,
|
| 2287 |
+
"eval_samples_per_second": 4.536,
|
| 2288 |
+
"eval_steps_per_second": 1.514,
|
| 2289 |
+
"step": 15200
|
| 2290 |
+
},
|
| 2291 |
+
{
|
| 2292 |
+
"epoch": 0.029546511261887724,
|
| 2293 |
+
"grad_norm": 0.8756581544876099,
|
| 2294 |
+
"learning_rate": 1.4773188111928626e-05,
|
| 2295 |
+
"loss": 1.9966,
|
| 2296 |
+
"step": 15300
|
| 2297 |
+
},
|
| 2298 |
+
{
|
| 2299 |
+
"epoch": 0.029546511261887724,
|
| 2300 |
+
"eval_loss": 2.04148530960083,
|
| 2301 |
+
"eval_runtime": 444.4494,
|
| 2302 |
+
"eval_samples_per_second": 4.504,
|
| 2303 |
+
"eval_steps_per_second": 1.503,
|
| 2304 |
+
"step": 15300
|
| 2305 |
+
},
|
| 2306 |
+
{
|
| 2307 |
+
"epoch": 0.0297396257145798,
|
| 2308 |
+
"grad_norm": 0.9758510589599609,
|
| 2309 |
+
"learning_rate": 1.4869744896973911e-05,
|
| 2310 |
+
"loss": 2.0051,
|
| 2311 |
+
"step": 15400
|
| 2312 |
+
},
|
| 2313 |
+
{
|
| 2314 |
+
"epoch": 0.0297396257145798,
|
| 2315 |
+
"eval_loss": 2.0388731956481934,
|
| 2316 |
+
"eval_runtime": 447.5472,
|
| 2317 |
+
"eval_samples_per_second": 4.473,
|
| 2318 |
+
"eval_steps_per_second": 1.493,
|
| 2319 |
+
"step": 15400
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.029932740167271876,
|
| 2323 |
+
"grad_norm": 0.8933822512626648,
|
| 2324 |
+
"learning_rate": 1.4966301682019196e-05,
|
| 2325 |
+
"loss": 2.004,
|
| 2326 |
+
"step": 15500
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 0.029932740167271876,
|
| 2330 |
+
"eval_loss": 2.0385355949401855,
|
| 2331 |
+
"eval_runtime": 448.7984,
|
| 2332 |
+
"eval_samples_per_second": 4.461,
|
| 2333 |
+
"eval_steps_per_second": 1.488,
|
| 2334 |
+
"step": 15500
|
| 2335 |
+
},
|
| 2336 |
+
{
|
| 2337 |
+
"epoch": 0.030125854619963952,
|
| 2338 |
+
"grad_norm": 1.0588964223861694,
|
| 2339 |
+
"learning_rate": 1.506285846706448e-05,
|
| 2340 |
+
"loss": 1.995,
|
| 2341 |
+
"step": 15600
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"epoch": 0.030125854619963952,
|
| 2345 |
+
"eval_loss": 2.0367350578308105,
|
| 2346 |
+
"eval_runtime": 446.8061,
|
| 2347 |
+
"eval_samples_per_second": 4.481,
|
| 2348 |
+
"eval_steps_per_second": 1.495,
|
| 2349 |
+
"step": 15600
|
| 2350 |
+
}
|
| 2351 |
+
],
|
| 2352 |
+
"logging_steps": 100,
|
| 2353 |
+
"max_steps": 517827,
|
| 2354 |
+
"num_input_tokens_seen": 0,
|
| 2355 |
+
"num_train_epochs": 1,
|
| 2356 |
+
"save_steps": 200,
|
| 2357 |
+
"stateful_callbacks": {
|
| 2358 |
+
"TrainerControl": {
|
| 2359 |
+
"args": {
|
| 2360 |
+
"should_epoch_stop": false,
|
| 2361 |
+
"should_evaluate": false,
|
| 2362 |
+
"should_log": false,
|
| 2363 |
+
"should_save": true,
|
| 2364 |
+
"should_training_stop": false
|
| 2365 |
+
},
|
| 2366 |
+
"attributes": {}
|
| 2367 |
+
}
|
| 2368 |
+
},
|
| 2369 |
+
"total_flos": 8.108193260524032e+18,
|
| 2370 |
+
"train_batch_size": 3,
|
| 2371 |
+
"trial_name": null,
|
| 2372 |
+
"trial_params": null
|
| 2373 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b3e2a69f44d5e3f0e5e1e4b03666c71840d278aaf14715e0446cdbc1e0da627b
|
| 3 |
+
size 5368
|