Upload folder using huggingface_hub
Browse files- .gitattributes +4 -0
- checkpoint-200/adapter_config.json +42 -0
- checkpoint-200/special_tokens_map.json +24 -0
- checkpoint-200/tokenizer.json +3 -0
- checkpoint-200/tokenizer_config.json +0 -0
- checkpoint-200/trainer_state.json +278 -0
- checkpoint-300/adapter_config.json +42 -0
- checkpoint-300/special_tokens_map.json +24 -0
- checkpoint-300/tokenizer.json +3 -0
- checkpoint-300/tokenizer_config.json +0 -0
- checkpoint-300/trainer_state.json +400 -0
- checkpoint-400/adapter_config.json +42 -0
- checkpoint-400/special_tokens_map.json +24 -0
- checkpoint-400/tokenizer.json +3 -0
- checkpoint-400/tokenizer_config.json +0 -0
- checkpoint-400/trainer_state.json +522 -0
- special_tokens_map.json +24 -0
- tokenizer.json +3 -0
- tokenizer_config.json +0 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,7 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
checkpoint-200/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
checkpoint-300/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 38 |
+
checkpoint-400/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 39 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
checkpoint-200/adapter_config.json
ADDED
|
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "/home/support/llm/Ministral-8B-Instruct-2410/",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"corda_config": null,
|
| 7 |
+
"eva_config": null,
|
| 8 |
+
"exclude_modules": null,
|
| 9 |
+
"fan_in_fan_out": false,
|
| 10 |
+
"inference_mode": true,
|
| 11 |
+
"init_lora_weights": true,
|
| 12 |
+
"layer_replication": null,
|
| 13 |
+
"layers_pattern": null,
|
| 14 |
+
"layers_to_transform": null,
|
| 15 |
+
"loftq_config": {},
|
| 16 |
+
"lora_alpha": 16,
|
| 17 |
+
"lora_bias": false,
|
| 18 |
+
"lora_dropout": 0.05,
|
| 19 |
+
"megatron_config": null,
|
| 20 |
+
"megatron_core": "megatron.core",
|
| 21 |
+
"modules_to_save": null,
|
| 22 |
+
"peft_type": "LORA",
|
| 23 |
+
"qalora_group_size": 16,
|
| 24 |
+
"r": 64,
|
| 25 |
+
"rank_pattern": {},
|
| 26 |
+
"revision": null,
|
| 27 |
+
"target_modules": [
|
| 28 |
+
"q_proj",
|
| 29 |
+
"down_proj",
|
| 30 |
+
"up_proj",
|
| 31 |
+
"v_proj",
|
| 32 |
+
"k_proj",
|
| 33 |
+
"o_proj",
|
| 34 |
+
"gate_proj"
|
| 35 |
+
],
|
| 36 |
+
"target_parameters": null,
|
| 37 |
+
"task_type": "CAUSAL_LM",
|
| 38 |
+
"trainable_token_indices": null,
|
| 39 |
+
"use_dora": false,
|
| 40 |
+
"use_qalora": false,
|
| 41 |
+
"use_rslora": false
|
| 42 |
+
}
|
checkpoint-200/special_tokens_map.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "</s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": "</s>",
|
| 17 |
+
"unk_token": {
|
| 18 |
+
"content": "<unk>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": false,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
}
|
| 24 |
+
}
|
checkpoint-200/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d7edbeaf20dd7f571b5dd1c54d9ace4f9b6299127cc7ba2afb14a6d51a4a79a4
|
| 3 |
+
size 17078136
|
checkpoint-200/tokenizer_config.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-200/trainer_state.json
ADDED
|
@@ -0,0 +1,278 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 200,
|
| 3 |
+
"best_metric": 0.9610546827316284,
|
| 4 |
+
"best_model_checkpoint": "./python-code-summarization-ministral-8B/checkpoint-200",
|
| 5 |
+
"epoch": 0.6274509803921569,
|
| 6 |
+
"eval_steps": 50,
|
| 7 |
+
"global_step": 200,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"entropy": 1.5161276400089263,
|
| 14 |
+
"epoch": 0.03137254901960784,
|
| 15 |
+
"grad_norm": 0.4193158745765686,
|
| 16 |
+
"learning_rate": 7.500000000000001e-06,
|
| 17 |
+
"loss": 1.7534,
|
| 18 |
+
"mean_token_accuracy": 0.5859726279973984,
|
| 19 |
+
"num_tokens": 40960.0,
|
| 20 |
+
"step": 10
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"entropy": 1.5331396609544754,
|
| 24 |
+
"epoch": 0.06274509803921569,
|
| 25 |
+
"grad_norm": 0.34798431396484375,
|
| 26 |
+
"learning_rate": 9.991971098713136e-06,
|
| 27 |
+
"loss": 1.7374,
|
| 28 |
+
"mean_token_accuracy": 0.5868768319487572,
|
| 29 |
+
"num_tokens": 81920.0,
|
| 30 |
+
"step": 20
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"entropy": 1.504272100329399,
|
| 34 |
+
"epoch": 0.09411764705882353,
|
| 35 |
+
"grad_norm": 0.29647937417030334,
|
| 36 |
+
"learning_rate": 9.952707929045018e-06,
|
| 37 |
+
"loss": 1.5992,
|
| 38 |
+
"mean_token_accuracy": 0.6027859210968017,
|
| 39 |
+
"num_tokens": 122880.0,
|
| 40 |
+
"step": 30
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"entropy": 1.4829266160726546,
|
| 44 |
+
"epoch": 0.12549019607843137,
|
| 45 |
+
"grad_norm": 0.35300737619400024,
|
| 46 |
+
"learning_rate": 9.88099273969759e-06,
|
| 47 |
+
"loss": 1.4991,
|
| 48 |
+
"mean_token_accuracy": 0.619208213686943,
|
| 49 |
+
"num_tokens": 163840.0,
|
| 50 |
+
"step": 40
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"entropy": 1.4123090922832489,
|
| 54 |
+
"epoch": 0.1568627450980392,
|
| 55 |
+
"grad_norm": 0.34853842854499817,
|
| 56 |
+
"learning_rate": 9.777295435775165e-06,
|
| 57 |
+
"loss": 1.4053,
|
| 58 |
+
"mean_token_accuracy": 0.6398732364177704,
|
| 59 |
+
"num_tokens": 204738.0,
|
| 60 |
+
"step": 50
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 0.1568627450980392,
|
| 64 |
+
"eval_entropy": 1.3041793243090312,
|
| 65 |
+
"eval_loss": 1.2978447675704956,
|
| 66 |
+
"eval_mean_token_accuracy": 0.6662148380279541,
|
| 67 |
+
"eval_num_tokens": 204738.0,
|
| 68 |
+
"eval_runtime": 55.6346,
|
| 69 |
+
"eval_samples_per_second": 4.044,
|
| 70 |
+
"eval_steps_per_second": 4.044,
|
| 71 |
+
"step": 50
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"entropy": 1.2701195865869521,
|
| 75 |
+
"epoch": 0.18823529411764706,
|
| 76 |
+
"grad_norm": 0.2973419427871704,
|
| 77 |
+
"learning_rate": 9.64229548131959e-06,
|
| 78 |
+
"loss": 1.2751,
|
| 79 |
+
"mean_token_accuracy": 0.6745112419128418,
|
| 80 |
+
"num_tokens": 245698.0,
|
| 81 |
+
"step": 60
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"entropy": 1.0886118918657304,
|
| 85 |
+
"epoch": 0.2196078431372549,
|
| 86 |
+
"grad_norm": 0.2847767770290375,
|
| 87 |
+
"learning_rate": 9.476877447204308e-06,
|
| 88 |
+
"loss": 1.1225,
|
| 89 |
+
"mean_token_accuracy": 0.7075268790125847,
|
| 90 |
+
"num_tokens": 286658.0,
|
| 91 |
+
"step": 70
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"entropy": 1.1023929134011268,
|
| 95 |
+
"epoch": 0.25098039215686274,
|
| 96 |
+
"grad_norm": 0.2556367814540863,
|
| 97 |
+
"learning_rate": 9.282125215090694e-06,
|
| 98 |
+
"loss": 1.1156,
|
| 99 |
+
"mean_token_accuracy": 0.7074291288852692,
|
| 100 |
+
"num_tokens": 327618.0,
|
| 101 |
+
"step": 80
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"entropy": 1.0319992125034332,
|
| 105 |
+
"epoch": 0.2823529411764706,
|
| 106 |
+
"grad_norm": 0.2943298816680908,
|
| 107 |
+
"learning_rate": 9.059314875424553e-06,
|
| 108 |
+
"loss": 1.0371,
|
| 109 |
+
"mean_token_accuracy": 0.7196480900049209,
|
| 110 |
+
"num_tokens": 368578.0,
|
| 111 |
+
"step": 90
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"entropy": 1.0712109550833702,
|
| 115 |
+
"epoch": 0.3137254901960784,
|
| 116 |
+
"grad_norm": 0.28212958574295044,
|
| 117 |
+
"learning_rate": 8.809906366007832e-06,
|
| 118 |
+
"loss": 1.0558,
|
| 119 |
+
"mean_token_accuracy": 0.716348971426487,
|
| 120 |
+
"num_tokens": 409538.0,
|
| 121 |
+
"step": 100
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.3137254901960784,
|
| 125 |
+
"eval_entropy": 1.0366067356533475,
|
| 126 |
+
"eval_loss": 1.0299878120422363,
|
| 127 |
+
"eval_mean_token_accuracy": 0.7208515254656473,
|
| 128 |
+
"eval_num_tokens": 409538.0,
|
| 129 |
+
"eval_runtime": 55.7692,
|
| 130 |
+
"eval_samples_per_second": 4.034,
|
| 131 |
+
"eval_steps_per_second": 4.034,
|
| 132 |
+
"step": 100
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"entropy": 1.0164774924516677,
|
| 136 |
+
"epoch": 0.34509803921568627,
|
| 137 |
+
"grad_norm": 0.28942403197288513,
|
| 138 |
+
"learning_rate": 8.535533905932739e-06,
|
| 139 |
+
"loss": 1.0079,
|
| 140 |
+
"mean_token_accuracy": 0.7282013669610023,
|
| 141 |
+
"num_tokens": 450498.0,
|
| 142 |
+
"step": 110
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"entropy": 1.0560581550002097,
|
| 146 |
+
"epoch": 0.3764705882352941,
|
| 147 |
+
"grad_norm": 0.2502649426460266,
|
| 148 |
+
"learning_rate": 8.237995287558801e-06,
|
| 149 |
+
"loss": 1.037,
|
| 150 |
+
"mean_token_accuracy": 0.7196969777345658,
|
| 151 |
+
"num_tokens": 491458.0,
|
| 152 |
+
"step": 120
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"entropy": 1.030363565683365,
|
| 156 |
+
"epoch": 0.40784313725490196,
|
| 157 |
+
"grad_norm": 0.27328935265541077,
|
| 158 |
+
"learning_rate": 7.919240096695877e-06,
|
| 159 |
+
"loss": 1.0113,
|
| 160 |
+
"mean_token_accuracy": 0.727834801375866,
|
| 161 |
+
"num_tokens": 532418.0,
|
| 162 |
+
"step": 130
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"entropy": 1.022090318799019,
|
| 166 |
+
"epoch": 0.4392156862745098,
|
| 167 |
+
"grad_norm": 0.2800595462322235,
|
| 168 |
+
"learning_rate": 7.58135693817893e-06,
|
| 169 |
+
"loss": 0.9984,
|
| 170 |
+
"mean_token_accuracy": 0.7261974602937699,
|
| 171 |
+
"num_tokens": 573378.0,
|
| 172 |
+
"step": 140
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"entropy": 1.0286607086658477,
|
| 176 |
+
"epoch": 0.47058823529411764,
|
| 177 |
+
"grad_norm": 0.29165375232696533,
|
| 178 |
+
"learning_rate": 7.226559750537462e-06,
|
| 179 |
+
"loss": 1.0059,
|
| 180 |
+
"mean_token_accuracy": 0.7231427162885666,
|
| 181 |
+
"num_tokens": 614338.0,
|
| 182 |
+
"step": 150
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.47058823529411764,
|
| 186 |
+
"eval_entropy": 1.0056122451358371,
|
| 187 |
+
"eval_loss": 0.983330488204956,
|
| 188 |
+
"eval_mean_token_accuracy": 0.7300966668128968,
|
| 189 |
+
"eval_num_tokens": 614338.0,
|
| 190 |
+
"eval_runtime": 55.6969,
|
| 191 |
+
"eval_samples_per_second": 4.04,
|
| 192 |
+
"eval_steps_per_second": 4.04,
|
| 193 |
+
"step": 150
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"entropy": 1.0160210341215135,
|
| 197 |
+
"epoch": 0.5019607843137255,
|
| 198 |
+
"grad_norm": 0.3049905598163605,
|
| 199 |
+
"learning_rate": 6.857173299431084e-06,
|
| 200 |
+
"loss": 0.9897,
|
| 201 |
+
"mean_token_accuracy": 0.7299120262265205,
|
| 202 |
+
"num_tokens": 655298.0,
|
| 203 |
+
"step": 160
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"entropy": 1.0173547983169555,
|
| 207 |
+
"epoch": 0.5333333333333333,
|
| 208 |
+
"grad_norm": 0.29565489292144775,
|
| 209 |
+
"learning_rate": 6.475617944903693e-06,
|
| 210 |
+
"loss": 1.0005,
|
| 211 |
+
"mean_token_accuracy": 0.7288611873984336,
|
| 212 |
+
"num_tokens": 696258.0,
|
| 213 |
+
"step": 170
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"entropy": 1.0144968718290328,
|
| 217 |
+
"epoch": 0.5647058823529412,
|
| 218 |
+
"grad_norm": 0.3220125436782837,
|
| 219 |
+
"learning_rate": 6.084393782267039e-06,
|
| 220 |
+
"loss": 0.9866,
|
| 221 |
+
"mean_token_accuracy": 0.7293499529361724,
|
| 222 |
+
"num_tokens": 737218.0,
|
| 223 |
+
"step": 180
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"entropy": 0.9772199034690857,
|
| 227 |
+
"epoch": 0.596078431372549,
|
| 228 |
+
"grad_norm": 0.32867270708084106,
|
| 229 |
+
"learning_rate": 5.686064260528577e-06,
|
| 230 |
+
"loss": 0.9618,
|
| 231 |
+
"mean_token_accuracy": 0.7332600191235542,
|
| 232 |
+
"num_tokens": 778178.0,
|
| 233 |
+
"step": 190
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"entropy": 0.9724952712655067,
|
| 237 |
+
"epoch": 0.6274509803921569,
|
| 238 |
+
"grad_norm": 0.3363828957080841,
|
| 239 |
+
"learning_rate": 5.283239385701882e-06,
|
| 240 |
+
"loss": 0.9558,
|
| 241 |
+
"mean_token_accuracy": 0.7365591391921044,
|
| 242 |
+
"num_tokens": 819138.0,
|
| 243 |
+
"step": 200
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 0.6274509803921569,
|
| 247 |
+
"eval_entropy": 0.9774409135182699,
|
| 248 |
+
"eval_loss": 0.9610546827316284,
|
| 249 |
+
"eval_mean_token_accuracy": 0.7337851632965936,
|
| 250 |
+
"eval_num_tokens": 819138.0,
|
| 251 |
+
"eval_runtime": 55.7878,
|
| 252 |
+
"eval_samples_per_second": 4.033,
|
| 253 |
+
"eval_steps_per_second": 4.033,
|
| 254 |
+
"step": 200
|
| 255 |
+
}
|
| 256 |
+
],
|
| 257 |
+
"logging_steps": 10,
|
| 258 |
+
"max_steps": 400,
|
| 259 |
+
"num_input_tokens_seen": 0,
|
| 260 |
+
"num_train_epochs": 2,
|
| 261 |
+
"save_steps": 100,
|
| 262 |
+
"stateful_callbacks": {
|
| 263 |
+
"TrainerControl": {
|
| 264 |
+
"args": {
|
| 265 |
+
"should_epoch_stop": false,
|
| 266 |
+
"should_evaluate": false,
|
| 267 |
+
"should_log": false,
|
| 268 |
+
"should_save": true,
|
| 269 |
+
"should_training_stop": false
|
| 270 |
+
},
|
| 271 |
+
"attributes": {}
|
| 272 |
+
}
|
| 273 |
+
},
|
| 274 |
+
"total_flos": 3.763541949957734e+16,
|
| 275 |
+
"train_batch_size": 1,
|
| 276 |
+
"trial_name": null,
|
| 277 |
+
"trial_params": null
|
| 278 |
+
}
|
checkpoint-300/adapter_config.json
ADDED
|
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "/home/support/llm/Ministral-8B-Instruct-2410/",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"corda_config": null,
|
| 7 |
+
"eva_config": null,
|
| 8 |
+
"exclude_modules": null,
|
| 9 |
+
"fan_in_fan_out": false,
|
| 10 |
+
"inference_mode": true,
|
| 11 |
+
"init_lora_weights": true,
|
| 12 |
+
"layer_replication": null,
|
| 13 |
+
"layers_pattern": null,
|
| 14 |
+
"layers_to_transform": null,
|
| 15 |
+
"loftq_config": {},
|
| 16 |
+
"lora_alpha": 16,
|
| 17 |
+
"lora_bias": false,
|
| 18 |
+
"lora_dropout": 0.05,
|
| 19 |
+
"megatron_config": null,
|
| 20 |
+
"megatron_core": "megatron.core",
|
| 21 |
+
"modules_to_save": null,
|
| 22 |
+
"peft_type": "LORA",
|
| 23 |
+
"qalora_group_size": 16,
|
| 24 |
+
"r": 64,
|
| 25 |
+
"rank_pattern": {},
|
| 26 |
+
"revision": null,
|
| 27 |
+
"target_modules": [
|
| 28 |
+
"q_proj",
|
| 29 |
+
"down_proj",
|
| 30 |
+
"up_proj",
|
| 31 |
+
"v_proj",
|
| 32 |
+
"k_proj",
|
| 33 |
+
"o_proj",
|
| 34 |
+
"gate_proj"
|
| 35 |
+
],
|
| 36 |
+
"target_parameters": null,
|
| 37 |
+
"task_type": "CAUSAL_LM",
|
| 38 |
+
"trainable_token_indices": null,
|
| 39 |
+
"use_dora": false,
|
| 40 |
+
"use_qalora": false,
|
| 41 |
+
"use_rslora": false
|
| 42 |
+
}
|
checkpoint-300/special_tokens_map.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "</s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": "</s>",
|
| 17 |
+
"unk_token": {
|
| 18 |
+
"content": "<unk>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": false,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
}
|
| 24 |
+
}
|
checkpoint-300/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d7edbeaf20dd7f571b5dd1c54d9ace4f9b6299127cc7ba2afb14a6d51a4a79a4
|
| 3 |
+
size 17078136
|
checkpoint-300/tokenizer_config.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-300/trainer_state.json
ADDED
|
@@ -0,0 +1,400 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 300,
|
| 3 |
+
"best_metric": 0.9439367651939392,
|
| 4 |
+
"best_model_checkpoint": "./python-code-summarization-ministral-8B/checkpoint-300",
|
| 5 |
+
"epoch": 0.9411764705882353,
|
| 6 |
+
"eval_steps": 50,
|
| 7 |
+
"global_step": 300,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"entropy": 1.5161276400089263,
|
| 14 |
+
"epoch": 0.03137254901960784,
|
| 15 |
+
"grad_norm": 0.4193158745765686,
|
| 16 |
+
"learning_rate": 7.500000000000001e-06,
|
| 17 |
+
"loss": 1.7534,
|
| 18 |
+
"mean_token_accuracy": 0.5859726279973984,
|
| 19 |
+
"num_tokens": 40960.0,
|
| 20 |
+
"step": 10
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"entropy": 1.5331396609544754,
|
| 24 |
+
"epoch": 0.06274509803921569,
|
| 25 |
+
"grad_norm": 0.34798431396484375,
|
| 26 |
+
"learning_rate": 9.991971098713136e-06,
|
| 27 |
+
"loss": 1.7374,
|
| 28 |
+
"mean_token_accuracy": 0.5868768319487572,
|
| 29 |
+
"num_tokens": 81920.0,
|
| 30 |
+
"step": 20
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"entropy": 1.504272100329399,
|
| 34 |
+
"epoch": 0.09411764705882353,
|
| 35 |
+
"grad_norm": 0.29647937417030334,
|
| 36 |
+
"learning_rate": 9.952707929045018e-06,
|
| 37 |
+
"loss": 1.5992,
|
| 38 |
+
"mean_token_accuracy": 0.6027859210968017,
|
| 39 |
+
"num_tokens": 122880.0,
|
| 40 |
+
"step": 30
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"entropy": 1.4829266160726546,
|
| 44 |
+
"epoch": 0.12549019607843137,
|
| 45 |
+
"grad_norm": 0.35300737619400024,
|
| 46 |
+
"learning_rate": 9.88099273969759e-06,
|
| 47 |
+
"loss": 1.4991,
|
| 48 |
+
"mean_token_accuracy": 0.619208213686943,
|
| 49 |
+
"num_tokens": 163840.0,
|
| 50 |
+
"step": 40
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"entropy": 1.4123090922832489,
|
| 54 |
+
"epoch": 0.1568627450980392,
|
| 55 |
+
"grad_norm": 0.34853842854499817,
|
| 56 |
+
"learning_rate": 9.777295435775165e-06,
|
| 57 |
+
"loss": 1.4053,
|
| 58 |
+
"mean_token_accuracy": 0.6398732364177704,
|
| 59 |
+
"num_tokens": 204738.0,
|
| 60 |
+
"step": 50
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 0.1568627450980392,
|
| 64 |
+
"eval_entropy": 1.3041793243090312,
|
| 65 |
+
"eval_loss": 1.2978447675704956,
|
| 66 |
+
"eval_mean_token_accuracy": 0.6662148380279541,
|
| 67 |
+
"eval_num_tokens": 204738.0,
|
| 68 |
+
"eval_runtime": 55.6346,
|
| 69 |
+
"eval_samples_per_second": 4.044,
|
| 70 |
+
"eval_steps_per_second": 4.044,
|
| 71 |
+
"step": 50
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"entropy": 1.2701195865869521,
|
| 75 |
+
"epoch": 0.18823529411764706,
|
| 76 |
+
"grad_norm": 0.2973419427871704,
|
| 77 |
+
"learning_rate": 9.64229548131959e-06,
|
| 78 |
+
"loss": 1.2751,
|
| 79 |
+
"mean_token_accuracy": 0.6745112419128418,
|
| 80 |
+
"num_tokens": 245698.0,
|
| 81 |
+
"step": 60
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"entropy": 1.0886118918657304,
|
| 85 |
+
"epoch": 0.2196078431372549,
|
| 86 |
+
"grad_norm": 0.2847767770290375,
|
| 87 |
+
"learning_rate": 9.476877447204308e-06,
|
| 88 |
+
"loss": 1.1225,
|
| 89 |
+
"mean_token_accuracy": 0.7075268790125847,
|
| 90 |
+
"num_tokens": 286658.0,
|
| 91 |
+
"step": 70
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"entropy": 1.1023929134011268,
|
| 95 |
+
"epoch": 0.25098039215686274,
|
| 96 |
+
"grad_norm": 0.2556367814540863,
|
| 97 |
+
"learning_rate": 9.282125215090694e-06,
|
| 98 |
+
"loss": 1.1156,
|
| 99 |
+
"mean_token_accuracy": 0.7074291288852692,
|
| 100 |
+
"num_tokens": 327618.0,
|
| 101 |
+
"step": 80
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"entropy": 1.0319992125034332,
|
| 105 |
+
"epoch": 0.2823529411764706,
|
| 106 |
+
"grad_norm": 0.2943298816680908,
|
| 107 |
+
"learning_rate": 9.059314875424553e-06,
|
| 108 |
+
"loss": 1.0371,
|
| 109 |
+
"mean_token_accuracy": 0.7196480900049209,
|
| 110 |
+
"num_tokens": 368578.0,
|
| 111 |
+
"step": 90
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"entropy": 1.0712109550833702,
|
| 115 |
+
"epoch": 0.3137254901960784,
|
| 116 |
+
"grad_norm": 0.28212958574295044,
|
| 117 |
+
"learning_rate": 8.809906366007832e-06,
|
| 118 |
+
"loss": 1.0558,
|
| 119 |
+
"mean_token_accuracy": 0.716348971426487,
|
| 120 |
+
"num_tokens": 409538.0,
|
| 121 |
+
"step": 100
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.3137254901960784,
|
| 125 |
+
"eval_entropy": 1.0366067356533475,
|
| 126 |
+
"eval_loss": 1.0299878120422363,
|
| 127 |
+
"eval_mean_token_accuracy": 0.7208515254656473,
|
| 128 |
+
"eval_num_tokens": 409538.0,
|
| 129 |
+
"eval_runtime": 55.7692,
|
| 130 |
+
"eval_samples_per_second": 4.034,
|
| 131 |
+
"eval_steps_per_second": 4.034,
|
| 132 |
+
"step": 100
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"entropy": 1.0164774924516677,
|
| 136 |
+
"epoch": 0.34509803921568627,
|
| 137 |
+
"grad_norm": 0.28942403197288513,
|
| 138 |
+
"learning_rate": 8.535533905932739e-06,
|
| 139 |
+
"loss": 1.0079,
|
| 140 |
+
"mean_token_accuracy": 0.7282013669610023,
|
| 141 |
+
"num_tokens": 450498.0,
|
| 142 |
+
"step": 110
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"entropy": 1.0560581550002097,
|
| 146 |
+
"epoch": 0.3764705882352941,
|
| 147 |
+
"grad_norm": 0.2502649426460266,
|
| 148 |
+
"learning_rate": 8.237995287558801e-06,
|
| 149 |
+
"loss": 1.037,
|
| 150 |
+
"mean_token_accuracy": 0.7196969777345658,
|
| 151 |
+
"num_tokens": 491458.0,
|
| 152 |
+
"step": 120
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"entropy": 1.030363565683365,
|
| 156 |
+
"epoch": 0.40784313725490196,
|
| 157 |
+
"grad_norm": 0.27328935265541077,
|
| 158 |
+
"learning_rate": 7.919240096695877e-06,
|
| 159 |
+
"loss": 1.0113,
|
| 160 |
+
"mean_token_accuracy": 0.727834801375866,
|
| 161 |
+
"num_tokens": 532418.0,
|
| 162 |
+
"step": 130
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"entropy": 1.022090318799019,
|
| 166 |
+
"epoch": 0.4392156862745098,
|
| 167 |
+
"grad_norm": 0.2800595462322235,
|
| 168 |
+
"learning_rate": 7.58135693817893e-06,
|
| 169 |
+
"loss": 0.9984,
|
| 170 |
+
"mean_token_accuracy": 0.7261974602937699,
|
| 171 |
+
"num_tokens": 573378.0,
|
| 172 |
+
"step": 140
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"entropy": 1.0286607086658477,
|
| 176 |
+
"epoch": 0.47058823529411764,
|
| 177 |
+
"grad_norm": 0.29165375232696533,
|
| 178 |
+
"learning_rate": 7.226559750537462e-06,
|
| 179 |
+
"loss": 1.0059,
|
| 180 |
+
"mean_token_accuracy": 0.7231427162885666,
|
| 181 |
+
"num_tokens": 614338.0,
|
| 182 |
+
"step": 150
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.47058823529411764,
|
| 186 |
+
"eval_entropy": 1.0056122451358371,
|
| 187 |
+
"eval_loss": 0.983330488204956,
|
| 188 |
+
"eval_mean_token_accuracy": 0.7300966668128968,
|
| 189 |
+
"eval_num_tokens": 614338.0,
|
| 190 |
+
"eval_runtime": 55.6969,
|
| 191 |
+
"eval_samples_per_second": 4.04,
|
| 192 |
+
"eval_steps_per_second": 4.04,
|
| 193 |
+
"step": 150
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"entropy": 1.0160210341215135,
|
| 197 |
+
"epoch": 0.5019607843137255,
|
| 198 |
+
"grad_norm": 0.3049905598163605,
|
| 199 |
+
"learning_rate": 6.857173299431084e-06,
|
| 200 |
+
"loss": 0.9897,
|
| 201 |
+
"mean_token_accuracy": 0.7299120262265205,
|
| 202 |
+
"num_tokens": 655298.0,
|
| 203 |
+
"step": 160
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"entropy": 1.0173547983169555,
|
| 207 |
+
"epoch": 0.5333333333333333,
|
| 208 |
+
"grad_norm": 0.29565489292144775,
|
| 209 |
+
"learning_rate": 6.475617944903693e-06,
|
| 210 |
+
"loss": 1.0005,
|
| 211 |
+
"mean_token_accuracy": 0.7288611873984336,
|
| 212 |
+
"num_tokens": 696258.0,
|
| 213 |
+
"step": 170
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"entropy": 1.0144968718290328,
|
| 217 |
+
"epoch": 0.5647058823529412,
|
| 218 |
+
"grad_norm": 0.3220125436782837,
|
| 219 |
+
"learning_rate": 6.084393782267039e-06,
|
| 220 |
+
"loss": 0.9866,
|
| 221 |
+
"mean_token_accuracy": 0.7293499529361724,
|
| 222 |
+
"num_tokens": 737218.0,
|
| 223 |
+
"step": 180
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"entropy": 0.9772199034690857,
|
| 227 |
+
"epoch": 0.596078431372549,
|
| 228 |
+
"grad_norm": 0.32867270708084106,
|
| 229 |
+
"learning_rate": 5.686064260528577e-06,
|
| 230 |
+
"loss": 0.9618,
|
| 231 |
+
"mean_token_accuracy": 0.7332600191235542,
|
| 232 |
+
"num_tokens": 778178.0,
|
| 233 |
+
"step": 190
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"entropy": 0.9724952712655067,
|
| 237 |
+
"epoch": 0.6274509803921569,
|
| 238 |
+
"grad_norm": 0.3363828957080841,
|
| 239 |
+
"learning_rate": 5.283239385701882e-06,
|
| 240 |
+
"loss": 0.9558,
|
| 241 |
+
"mean_token_accuracy": 0.7365591391921044,
|
| 242 |
+
"num_tokens": 819138.0,
|
| 243 |
+
"step": 200
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 0.6274509803921569,
|
| 247 |
+
"eval_entropy": 0.9774409135182699,
|
| 248 |
+
"eval_loss": 0.9610546827316284,
|
| 249 |
+
"eval_mean_token_accuracy": 0.7337851632965936,
|
| 250 |
+
"eval_num_tokens": 819138.0,
|
| 251 |
+
"eval_runtime": 55.7878,
|
| 252 |
+
"eval_samples_per_second": 4.033,
|
| 253 |
+
"eval_steps_per_second": 4.033,
|
| 254 |
+
"step": 200
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"entropy": 0.98491101115942,
|
| 258 |
+
"epoch": 0.6588235294117647,
|
| 259 |
+
"grad_norm": 0.3083605170249939,
|
| 260 |
+
"learning_rate": 4.8785586190579355e-06,
|
| 261 |
+
"loss": 0.9618,
|
| 262 |
+
"mean_token_accuracy": 0.7304252177476883,
|
| 263 |
+
"num_tokens": 860098.0,
|
| 264 |
+
"step": 210
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"entropy": 1.0157848447561264,
|
| 268 |
+
"epoch": 0.6901960784313725,
|
| 269 |
+
"grad_norm": 0.30706891417503357,
|
| 270 |
+
"learning_rate": 4.474673582374381e-06,
|
| 271 |
+
"loss": 1.0052,
|
| 272 |
+
"mean_token_accuracy": 0.7238514170050621,
|
| 273 |
+
"num_tokens": 901058.0,
|
| 274 |
+
"step": 220
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"entropy": 0.9612855270504952,
|
| 278 |
+
"epoch": 0.7215686274509804,
|
| 279 |
+
"grad_norm": 0.3250674605369568,
|
| 280 |
+
"learning_rate": 4.074230683504541e-06,
|
| 281 |
+
"loss": 0.9545,
|
| 282 |
+
"mean_token_accuracy": 0.7387096747756005,
|
| 283 |
+
"num_tokens": 942018.0,
|
| 284 |
+
"step": 230
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"entropy": 0.9811745539307595,
|
| 288 |
+
"epoch": 0.7529411764705882,
|
| 289 |
+
"grad_norm": 0.3410372734069824,
|
| 290 |
+
"learning_rate": 3.679853776110035e-06,
|
| 291 |
+
"loss": 0.9529,
|
| 292 |
+
"mean_token_accuracy": 0.7300586521625518,
|
| 293 |
+
"num_tokens": 982978.0,
|
| 294 |
+
"step": 240
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"entropy": 0.9820658221840859,
|
| 298 |
+
"epoch": 0.7843137254901961,
|
| 299 |
+
"grad_norm": 0.3213532567024231,
|
| 300 |
+
"learning_rate": 3.2941269671770193e-06,
|
| 301 |
+
"loss": 0.9632,
|
| 302 |
+
"mean_token_accuracy": 0.7317448645830155,
|
| 303 |
+
"num_tokens": 1023938.0,
|
| 304 |
+
"step": 250
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.7843137254901961,
|
| 308 |
+
"eval_entropy": 0.9663468877474467,
|
| 309 |
+
"eval_loss": 0.9494678974151611,
|
| 310 |
+
"eval_mean_token_accuracy": 0.7356924083497789,
|
| 311 |
+
"eval_num_tokens": 1023938.0,
|
| 312 |
+
"eval_runtime": 55.6726,
|
| 313 |
+
"eval_samples_per_second": 4.041,
|
| 314 |
+
"eval_steps_per_second": 4.041,
|
| 315 |
+
"step": 250
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"entropy": 0.9845806032419204,
|
| 319 |
+
"epoch": 0.8156862745098039,
|
| 320 |
+
"grad_norm": 0.3576700687408447,
|
| 321 |
+
"learning_rate": 2.9195776849677037e-06,
|
| 322 |
+
"loss": 0.9711,
|
| 323 |
+
"mean_token_accuracy": 0.7322091892361641,
|
| 324 |
+
"num_tokens": 1064898.0,
|
| 325 |
+
"step": 260
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"entropy": 0.9538720920681953,
|
| 329 |
+
"epoch": 0.8470588235294118,
|
| 330 |
+
"grad_norm": 0.34291872382164,
|
| 331 |
+
"learning_rate": 2.5586601183523485e-06,
|
| 332 |
+
"loss": 0.934,
|
| 333 |
+
"mean_token_accuracy": 0.7407869040966034,
|
| 334 |
+
"num_tokens": 1105858.0,
|
| 335 |
+
"step": 270
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"entropy": 0.931637741625309,
|
| 339 |
+
"epoch": 0.8784313725490196,
|
| 340 |
+
"grad_norm": 0.3526791036128998,
|
| 341 |
+
"learning_rate": 2.213739136033533e-06,
|
| 342 |
+
"loss": 0.9127,
|
| 343 |
+
"mean_token_accuracy": 0.745845553278923,
|
| 344 |
+
"num_tokens": 1146818.0,
|
| 345 |
+
"step": 280
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"entropy": 0.9876263022422791,
|
| 349 |
+
"epoch": 0.9098039215686274,
|
| 350 |
+
"grad_norm": 0.36476704478263855,
|
| 351 |
+
"learning_rate": 1.8870747910300064e-06,
|
| 352 |
+
"loss": 0.9755,
|
| 353 |
+
"mean_token_accuracy": 0.7316226810216904,
|
| 354 |
+
"num_tokens": 1187778.0,
|
| 355 |
+
"step": 290
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"entropy": 0.9182186275720596,
|
| 359 |
+
"epoch": 0.9411764705882353,
|
| 360 |
+
"grad_norm": 0.3551705777645111,
|
| 361 |
+
"learning_rate": 1.5808075119526323e-06,
|
| 362 |
+
"loss": 0.8992,
|
| 363 |
+
"mean_token_accuracy": 0.7443059593439102,
|
| 364 |
+
"num_tokens": 1228738.0,
|
| 365 |
+
"step": 300
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 0.9411764705882353,
|
| 369 |
+
"eval_entropy": 0.960735419591268,
|
| 370 |
+
"eval_loss": 0.9439367651939392,
|
| 371 |
+
"eval_mean_token_accuracy": 0.7366395137045119,
|
| 372 |
+
"eval_num_tokens": 1228738.0,
|
| 373 |
+
"eval_runtime": 55.8238,
|
| 374 |
+
"eval_samples_per_second": 4.031,
|
| 375 |
+
"eval_steps_per_second": 4.031,
|
| 376 |
+
"step": 300
|
| 377 |
+
}
|
| 378 |
+
],
|
| 379 |
+
"logging_steps": 10,
|
| 380 |
+
"max_steps": 400,
|
| 381 |
+
"num_input_tokens_seen": 0,
|
| 382 |
+
"num_train_epochs": 2,
|
| 383 |
+
"save_steps": 100,
|
| 384 |
+
"stateful_callbacks": {
|
| 385 |
+
"TrainerControl": {
|
| 386 |
+
"args": {
|
| 387 |
+
"should_epoch_stop": false,
|
| 388 |
+
"should_evaluate": false,
|
| 389 |
+
"should_log": false,
|
| 390 |
+
"should_save": true,
|
| 391 |
+
"should_training_stop": false
|
| 392 |
+
},
|
| 393 |
+
"attributes": {}
|
| 394 |
+
}
|
| 395 |
+
},
|
| 396 |
+
"total_flos": 5.645455354906214e+16,
|
| 397 |
+
"train_batch_size": 1,
|
| 398 |
+
"trial_name": null,
|
| 399 |
+
"trial_params": null
|
| 400 |
+
}
|
checkpoint-400/adapter_config.json
ADDED
|
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "/home/support/llm/Ministral-8B-Instruct-2410/",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"corda_config": null,
|
| 7 |
+
"eva_config": null,
|
| 8 |
+
"exclude_modules": null,
|
| 9 |
+
"fan_in_fan_out": false,
|
| 10 |
+
"inference_mode": true,
|
| 11 |
+
"init_lora_weights": true,
|
| 12 |
+
"layer_replication": null,
|
| 13 |
+
"layers_pattern": null,
|
| 14 |
+
"layers_to_transform": null,
|
| 15 |
+
"loftq_config": {},
|
| 16 |
+
"lora_alpha": 16,
|
| 17 |
+
"lora_bias": false,
|
| 18 |
+
"lora_dropout": 0.05,
|
| 19 |
+
"megatron_config": null,
|
| 20 |
+
"megatron_core": "megatron.core",
|
| 21 |
+
"modules_to_save": null,
|
| 22 |
+
"peft_type": "LORA",
|
| 23 |
+
"qalora_group_size": 16,
|
| 24 |
+
"r": 64,
|
| 25 |
+
"rank_pattern": {},
|
| 26 |
+
"revision": null,
|
| 27 |
+
"target_modules": [
|
| 28 |
+
"q_proj",
|
| 29 |
+
"down_proj",
|
| 30 |
+
"up_proj",
|
| 31 |
+
"v_proj",
|
| 32 |
+
"k_proj",
|
| 33 |
+
"o_proj",
|
| 34 |
+
"gate_proj"
|
| 35 |
+
],
|
| 36 |
+
"target_parameters": null,
|
| 37 |
+
"task_type": "CAUSAL_LM",
|
| 38 |
+
"trainable_token_indices": null,
|
| 39 |
+
"use_dora": false,
|
| 40 |
+
"use_qalora": false,
|
| 41 |
+
"use_rslora": false
|
| 42 |
+
}
|
checkpoint-400/special_tokens_map.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "</s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": "</s>",
|
| 17 |
+
"unk_token": {
|
| 18 |
+
"content": "<unk>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": false,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
}
|
| 24 |
+
}
|
checkpoint-400/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d7edbeaf20dd7f571b5dd1c54d9ace4f9b6299127cc7ba2afb14a6d51a4a79a4
|
| 3 |
+
size 17078136
|
checkpoint-400/tokenizer_config.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-400/trainer_state.json
ADDED
|
@@ -0,0 +1,522 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 400,
|
| 3 |
+
"best_metric": 0.9414717555046082,
|
| 4 |
+
"best_model_checkpoint": "./python-code-summarization-ministral-8B/checkpoint-400",
|
| 5 |
+
"epoch": 1.2541176470588236,
|
| 6 |
+
"eval_steps": 50,
|
| 7 |
+
"global_step": 400,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"entropy": 1.5161276400089263,
|
| 14 |
+
"epoch": 0.03137254901960784,
|
| 15 |
+
"grad_norm": 0.4193158745765686,
|
| 16 |
+
"learning_rate": 7.500000000000001e-06,
|
| 17 |
+
"loss": 1.7534,
|
| 18 |
+
"mean_token_accuracy": 0.5859726279973984,
|
| 19 |
+
"num_tokens": 40960.0,
|
| 20 |
+
"step": 10
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"entropy": 1.5331396609544754,
|
| 24 |
+
"epoch": 0.06274509803921569,
|
| 25 |
+
"grad_norm": 0.34798431396484375,
|
| 26 |
+
"learning_rate": 9.991971098713136e-06,
|
| 27 |
+
"loss": 1.7374,
|
| 28 |
+
"mean_token_accuracy": 0.5868768319487572,
|
| 29 |
+
"num_tokens": 81920.0,
|
| 30 |
+
"step": 20
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"entropy": 1.504272100329399,
|
| 34 |
+
"epoch": 0.09411764705882353,
|
| 35 |
+
"grad_norm": 0.29647937417030334,
|
| 36 |
+
"learning_rate": 9.952707929045018e-06,
|
| 37 |
+
"loss": 1.5992,
|
| 38 |
+
"mean_token_accuracy": 0.6027859210968017,
|
| 39 |
+
"num_tokens": 122880.0,
|
| 40 |
+
"step": 30
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"entropy": 1.4829266160726546,
|
| 44 |
+
"epoch": 0.12549019607843137,
|
| 45 |
+
"grad_norm": 0.35300737619400024,
|
| 46 |
+
"learning_rate": 9.88099273969759e-06,
|
| 47 |
+
"loss": 1.4991,
|
| 48 |
+
"mean_token_accuracy": 0.619208213686943,
|
| 49 |
+
"num_tokens": 163840.0,
|
| 50 |
+
"step": 40
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"entropy": 1.4123090922832489,
|
| 54 |
+
"epoch": 0.1568627450980392,
|
| 55 |
+
"grad_norm": 0.34853842854499817,
|
| 56 |
+
"learning_rate": 9.777295435775165e-06,
|
| 57 |
+
"loss": 1.4053,
|
| 58 |
+
"mean_token_accuracy": 0.6398732364177704,
|
| 59 |
+
"num_tokens": 204738.0,
|
| 60 |
+
"step": 50
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 0.1568627450980392,
|
| 64 |
+
"eval_entropy": 1.3041793243090312,
|
| 65 |
+
"eval_loss": 1.2978447675704956,
|
| 66 |
+
"eval_mean_token_accuracy": 0.6662148380279541,
|
| 67 |
+
"eval_num_tokens": 204738.0,
|
| 68 |
+
"eval_runtime": 55.6346,
|
| 69 |
+
"eval_samples_per_second": 4.044,
|
| 70 |
+
"eval_steps_per_second": 4.044,
|
| 71 |
+
"step": 50
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"entropy": 1.2701195865869521,
|
| 75 |
+
"epoch": 0.18823529411764706,
|
| 76 |
+
"grad_norm": 0.2973419427871704,
|
| 77 |
+
"learning_rate": 9.64229548131959e-06,
|
| 78 |
+
"loss": 1.2751,
|
| 79 |
+
"mean_token_accuracy": 0.6745112419128418,
|
| 80 |
+
"num_tokens": 245698.0,
|
| 81 |
+
"step": 60
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"entropy": 1.0886118918657304,
|
| 85 |
+
"epoch": 0.2196078431372549,
|
| 86 |
+
"grad_norm": 0.2847767770290375,
|
| 87 |
+
"learning_rate": 9.476877447204308e-06,
|
| 88 |
+
"loss": 1.1225,
|
| 89 |
+
"mean_token_accuracy": 0.7075268790125847,
|
| 90 |
+
"num_tokens": 286658.0,
|
| 91 |
+
"step": 70
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"entropy": 1.1023929134011268,
|
| 95 |
+
"epoch": 0.25098039215686274,
|
| 96 |
+
"grad_norm": 0.2556367814540863,
|
| 97 |
+
"learning_rate": 9.282125215090694e-06,
|
| 98 |
+
"loss": 1.1156,
|
| 99 |
+
"mean_token_accuracy": 0.7074291288852692,
|
| 100 |
+
"num_tokens": 327618.0,
|
| 101 |
+
"step": 80
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"entropy": 1.0319992125034332,
|
| 105 |
+
"epoch": 0.2823529411764706,
|
| 106 |
+
"grad_norm": 0.2943298816680908,
|
| 107 |
+
"learning_rate": 9.059314875424553e-06,
|
| 108 |
+
"loss": 1.0371,
|
| 109 |
+
"mean_token_accuracy": 0.7196480900049209,
|
| 110 |
+
"num_tokens": 368578.0,
|
| 111 |
+
"step": 90
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"entropy": 1.0712109550833702,
|
| 115 |
+
"epoch": 0.3137254901960784,
|
| 116 |
+
"grad_norm": 0.28212958574295044,
|
| 117 |
+
"learning_rate": 8.809906366007832e-06,
|
| 118 |
+
"loss": 1.0558,
|
| 119 |
+
"mean_token_accuracy": 0.716348971426487,
|
| 120 |
+
"num_tokens": 409538.0,
|
| 121 |
+
"step": 100
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.3137254901960784,
|
| 125 |
+
"eval_entropy": 1.0366067356533475,
|
| 126 |
+
"eval_loss": 1.0299878120422363,
|
| 127 |
+
"eval_mean_token_accuracy": 0.7208515254656473,
|
| 128 |
+
"eval_num_tokens": 409538.0,
|
| 129 |
+
"eval_runtime": 55.7692,
|
| 130 |
+
"eval_samples_per_second": 4.034,
|
| 131 |
+
"eval_steps_per_second": 4.034,
|
| 132 |
+
"step": 100
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"entropy": 1.0164774924516677,
|
| 136 |
+
"epoch": 0.34509803921568627,
|
| 137 |
+
"grad_norm": 0.28942403197288513,
|
| 138 |
+
"learning_rate": 8.535533905932739e-06,
|
| 139 |
+
"loss": 1.0079,
|
| 140 |
+
"mean_token_accuracy": 0.7282013669610023,
|
| 141 |
+
"num_tokens": 450498.0,
|
| 142 |
+
"step": 110
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"entropy": 1.0560581550002097,
|
| 146 |
+
"epoch": 0.3764705882352941,
|
| 147 |
+
"grad_norm": 0.2502649426460266,
|
| 148 |
+
"learning_rate": 8.237995287558801e-06,
|
| 149 |
+
"loss": 1.037,
|
| 150 |
+
"mean_token_accuracy": 0.7196969777345658,
|
| 151 |
+
"num_tokens": 491458.0,
|
| 152 |
+
"step": 120
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"entropy": 1.030363565683365,
|
| 156 |
+
"epoch": 0.40784313725490196,
|
| 157 |
+
"grad_norm": 0.27328935265541077,
|
| 158 |
+
"learning_rate": 7.919240096695877e-06,
|
| 159 |
+
"loss": 1.0113,
|
| 160 |
+
"mean_token_accuracy": 0.727834801375866,
|
| 161 |
+
"num_tokens": 532418.0,
|
| 162 |
+
"step": 130
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"entropy": 1.022090318799019,
|
| 166 |
+
"epoch": 0.4392156862745098,
|
| 167 |
+
"grad_norm": 0.2800595462322235,
|
| 168 |
+
"learning_rate": 7.58135693817893e-06,
|
| 169 |
+
"loss": 0.9984,
|
| 170 |
+
"mean_token_accuracy": 0.7261974602937699,
|
| 171 |
+
"num_tokens": 573378.0,
|
| 172 |
+
"step": 140
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"entropy": 1.0286607086658477,
|
| 176 |
+
"epoch": 0.47058823529411764,
|
| 177 |
+
"grad_norm": 0.29165375232696533,
|
| 178 |
+
"learning_rate": 7.226559750537462e-06,
|
| 179 |
+
"loss": 1.0059,
|
| 180 |
+
"mean_token_accuracy": 0.7231427162885666,
|
| 181 |
+
"num_tokens": 614338.0,
|
| 182 |
+
"step": 150
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.47058823529411764,
|
| 186 |
+
"eval_entropy": 1.0056122451358371,
|
| 187 |
+
"eval_loss": 0.983330488204956,
|
| 188 |
+
"eval_mean_token_accuracy": 0.7300966668128968,
|
| 189 |
+
"eval_num_tokens": 614338.0,
|
| 190 |
+
"eval_runtime": 55.6969,
|
| 191 |
+
"eval_samples_per_second": 4.04,
|
| 192 |
+
"eval_steps_per_second": 4.04,
|
| 193 |
+
"step": 150
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"entropy": 1.0160210341215135,
|
| 197 |
+
"epoch": 0.5019607843137255,
|
| 198 |
+
"grad_norm": 0.3049905598163605,
|
| 199 |
+
"learning_rate": 6.857173299431084e-06,
|
| 200 |
+
"loss": 0.9897,
|
| 201 |
+
"mean_token_accuracy": 0.7299120262265205,
|
| 202 |
+
"num_tokens": 655298.0,
|
| 203 |
+
"step": 160
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"entropy": 1.0173547983169555,
|
| 207 |
+
"epoch": 0.5333333333333333,
|
| 208 |
+
"grad_norm": 0.29565489292144775,
|
| 209 |
+
"learning_rate": 6.475617944903693e-06,
|
| 210 |
+
"loss": 1.0005,
|
| 211 |
+
"mean_token_accuracy": 0.7288611873984336,
|
| 212 |
+
"num_tokens": 696258.0,
|
| 213 |
+
"step": 170
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"entropy": 1.0144968718290328,
|
| 217 |
+
"epoch": 0.5647058823529412,
|
| 218 |
+
"grad_norm": 0.3220125436782837,
|
| 219 |
+
"learning_rate": 6.084393782267039e-06,
|
| 220 |
+
"loss": 0.9866,
|
| 221 |
+
"mean_token_accuracy": 0.7293499529361724,
|
| 222 |
+
"num_tokens": 737218.0,
|
| 223 |
+
"step": 180
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"entropy": 0.9772199034690857,
|
| 227 |
+
"epoch": 0.596078431372549,
|
| 228 |
+
"grad_norm": 0.32867270708084106,
|
| 229 |
+
"learning_rate": 5.686064260528577e-06,
|
| 230 |
+
"loss": 0.9618,
|
| 231 |
+
"mean_token_accuracy": 0.7332600191235542,
|
| 232 |
+
"num_tokens": 778178.0,
|
| 233 |
+
"step": 190
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"entropy": 0.9724952712655067,
|
| 237 |
+
"epoch": 0.6274509803921569,
|
| 238 |
+
"grad_norm": 0.3363828957080841,
|
| 239 |
+
"learning_rate": 5.283239385701882e-06,
|
| 240 |
+
"loss": 0.9558,
|
| 241 |
+
"mean_token_accuracy": 0.7365591391921044,
|
| 242 |
+
"num_tokens": 819138.0,
|
| 243 |
+
"step": 200
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 0.6274509803921569,
|
| 247 |
+
"eval_entropy": 0.9774409135182699,
|
| 248 |
+
"eval_loss": 0.9610546827316284,
|
| 249 |
+
"eval_mean_token_accuracy": 0.7337851632965936,
|
| 250 |
+
"eval_num_tokens": 819138.0,
|
| 251 |
+
"eval_runtime": 55.7878,
|
| 252 |
+
"eval_samples_per_second": 4.033,
|
| 253 |
+
"eval_steps_per_second": 4.033,
|
| 254 |
+
"step": 200
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"entropy": 0.98491101115942,
|
| 258 |
+
"epoch": 0.6588235294117647,
|
| 259 |
+
"grad_norm": 0.3083605170249939,
|
| 260 |
+
"learning_rate": 4.8785586190579355e-06,
|
| 261 |
+
"loss": 0.9618,
|
| 262 |
+
"mean_token_accuracy": 0.7304252177476883,
|
| 263 |
+
"num_tokens": 860098.0,
|
| 264 |
+
"step": 210
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"entropy": 1.0157848447561264,
|
| 268 |
+
"epoch": 0.6901960784313725,
|
| 269 |
+
"grad_norm": 0.30706891417503357,
|
| 270 |
+
"learning_rate": 4.474673582374381e-06,
|
| 271 |
+
"loss": 1.0052,
|
| 272 |
+
"mean_token_accuracy": 0.7238514170050621,
|
| 273 |
+
"num_tokens": 901058.0,
|
| 274 |
+
"step": 220
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"entropy": 0.9612855270504952,
|
| 278 |
+
"epoch": 0.7215686274509804,
|
| 279 |
+
"grad_norm": 0.3250674605369568,
|
| 280 |
+
"learning_rate": 4.074230683504541e-06,
|
| 281 |
+
"loss": 0.9545,
|
| 282 |
+
"mean_token_accuracy": 0.7387096747756005,
|
| 283 |
+
"num_tokens": 942018.0,
|
| 284 |
+
"step": 230
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"entropy": 0.9811745539307595,
|
| 288 |
+
"epoch": 0.7529411764705882,
|
| 289 |
+
"grad_norm": 0.3410372734069824,
|
| 290 |
+
"learning_rate": 3.679853776110035e-06,
|
| 291 |
+
"loss": 0.9529,
|
| 292 |
+
"mean_token_accuracy": 0.7300586521625518,
|
| 293 |
+
"num_tokens": 982978.0,
|
| 294 |
+
"step": 240
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"entropy": 0.9820658221840859,
|
| 298 |
+
"epoch": 0.7843137254901961,
|
| 299 |
+
"grad_norm": 0.3213532567024231,
|
| 300 |
+
"learning_rate": 3.2941269671770193e-06,
|
| 301 |
+
"loss": 0.9632,
|
| 302 |
+
"mean_token_accuracy": 0.7317448645830155,
|
| 303 |
+
"num_tokens": 1023938.0,
|
| 304 |
+
"step": 250
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.7843137254901961,
|
| 308 |
+
"eval_entropy": 0.9663468877474467,
|
| 309 |
+
"eval_loss": 0.9494678974151611,
|
| 310 |
+
"eval_mean_token_accuracy": 0.7356924083497789,
|
| 311 |
+
"eval_num_tokens": 1023938.0,
|
| 312 |
+
"eval_runtime": 55.6726,
|
| 313 |
+
"eval_samples_per_second": 4.041,
|
| 314 |
+
"eval_steps_per_second": 4.041,
|
| 315 |
+
"step": 250
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"entropy": 0.9845806032419204,
|
| 319 |
+
"epoch": 0.8156862745098039,
|
| 320 |
+
"grad_norm": 0.3576700687408447,
|
| 321 |
+
"learning_rate": 2.9195776849677037e-06,
|
| 322 |
+
"loss": 0.9711,
|
| 323 |
+
"mean_token_accuracy": 0.7322091892361641,
|
| 324 |
+
"num_tokens": 1064898.0,
|
| 325 |
+
"step": 260
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"entropy": 0.9538720920681953,
|
| 329 |
+
"epoch": 0.8470588235294118,
|
| 330 |
+
"grad_norm": 0.34291872382164,
|
| 331 |
+
"learning_rate": 2.5586601183523485e-06,
|
| 332 |
+
"loss": 0.934,
|
| 333 |
+
"mean_token_accuracy": 0.7407869040966034,
|
| 334 |
+
"num_tokens": 1105858.0,
|
| 335 |
+
"step": 270
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"entropy": 0.931637741625309,
|
| 339 |
+
"epoch": 0.8784313725490196,
|
| 340 |
+
"grad_norm": 0.3526791036128998,
|
| 341 |
+
"learning_rate": 2.213739136033533e-06,
|
| 342 |
+
"loss": 0.9127,
|
| 343 |
+
"mean_token_accuracy": 0.745845553278923,
|
| 344 |
+
"num_tokens": 1146818.0,
|
| 345 |
+
"step": 280
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"entropy": 0.9876263022422791,
|
| 349 |
+
"epoch": 0.9098039215686274,
|
| 350 |
+
"grad_norm": 0.36476704478263855,
|
| 351 |
+
"learning_rate": 1.8870747910300064e-06,
|
| 352 |
+
"loss": 0.9755,
|
| 353 |
+
"mean_token_accuracy": 0.7316226810216904,
|
| 354 |
+
"num_tokens": 1187778.0,
|
| 355 |
+
"step": 290
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"entropy": 0.9182186275720596,
|
| 359 |
+
"epoch": 0.9411764705882353,
|
| 360 |
+
"grad_norm": 0.3551705777645111,
|
| 361 |
+
"learning_rate": 1.5808075119526323e-06,
|
| 362 |
+
"loss": 0.8992,
|
| 363 |
+
"mean_token_accuracy": 0.7443059593439102,
|
| 364 |
+
"num_tokens": 1228738.0,
|
| 365 |
+
"step": 300
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 0.9411764705882353,
|
| 369 |
+
"eval_entropy": 0.960735419591268,
|
| 370 |
+
"eval_loss": 0.9439367651939392,
|
| 371 |
+
"eval_mean_token_accuracy": 0.7366395137045119,
|
| 372 |
+
"eval_num_tokens": 1228738.0,
|
| 373 |
+
"eval_runtime": 55.8238,
|
| 374 |
+
"eval_samples_per_second": 4.031,
|
| 375 |
+
"eval_steps_per_second": 4.031,
|
| 376 |
+
"step": 300
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"entropy": 0.9542466774582863,
|
| 380 |
+
"epoch": 0.9725490196078431,
|
| 381 |
+
"grad_norm": 0.3290831744670868,
|
| 382 |
+
"learning_rate": 1.2969440781048015e-06,
|
| 383 |
+
"loss": 0.9373,
|
| 384 |
+
"mean_token_accuracy": 0.7406158342957496,
|
| 385 |
+
"num_tokens": 1269698.0,
|
| 386 |
+
"step": 310
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"entropy": 0.992382885553898,
|
| 390 |
+
"epoch": 1.0031372549019608,
|
| 391 |
+
"grad_norm": 0.35219520330429077,
|
| 392 |
+
"learning_rate": 1.0373444703037644e-06,
|
| 393 |
+
"loss": 0.9847,
|
| 394 |
+
"mean_token_accuracy": 0.7291275018300766,
|
| 395 |
+
"num_tokens": 1309634.0,
|
| 396 |
+
"step": 320
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"entropy": 0.9549886897206307,
|
| 400 |
+
"epoch": 1.0345098039215685,
|
| 401 |
+
"grad_norm": 0.33679652214050293,
|
| 402 |
+
"learning_rate": 8.037096835812885e-07,
|
| 403 |
+
"loss": 0.9334,
|
| 404 |
+
"mean_token_accuracy": 0.7392717510461807,
|
| 405 |
+
"num_tokens": 1350594.0,
|
| 406 |
+
"step": 330
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"entropy": 0.9770316660404206,
|
| 410 |
+
"epoch": 1.0658823529411765,
|
| 411 |
+
"grad_norm": 0.3329678475856781,
|
| 412 |
+
"learning_rate": 5.97570581619446e-07,
|
| 413 |
+
"loss": 0.9528,
|
| 414 |
+
"mean_token_accuracy": 0.7311094805598259,
|
| 415 |
+
"num_tokens": 1391554.0,
|
| 416 |
+
"step": 340
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"entropy": 0.9595620423555374,
|
| 420 |
+
"epoch": 1.0972549019607842,
|
| 421 |
+
"grad_norm": 0.3248395323753357,
|
| 422 |
+
"learning_rate": 4.2027786595149555e-07,
|
| 423 |
+
"loss": 0.9332,
|
| 424 |
+
"mean_token_accuracy": 0.7348729267716407,
|
| 425 |
+
"num_tokens": 1432514.0,
|
| 426 |
+
"step": 350
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 1.0972549019607842,
|
| 430 |
+
"eval_entropy": 0.9590845820638868,
|
| 431 |
+
"eval_loss": 0.9417916536331177,
|
| 432 |
+
"eval_mean_token_accuracy": 0.737252090771993,
|
| 433 |
+
"eval_num_tokens": 1432514.0,
|
| 434 |
+
"eval_runtime": 55.6945,
|
| 435 |
+
"eval_samples_per_second": 4.04,
|
| 436 |
+
"eval_steps_per_second": 4.04,
|
| 437 |
+
"step": 350
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"entropy": 0.9660995692014694,
|
| 441 |
+
"epoch": 1.1286274509803922,
|
| 442 |
+
"grad_norm": 0.3455366790294647,
|
| 443 |
+
"learning_rate": 2.7299322565344956e-07,
|
| 444 |
+
"loss": 0.9624,
|
| 445 |
+
"mean_token_accuracy": 0.7324291303753853,
|
| 446 |
+
"num_tokens": 1473474.0,
|
| 447 |
+
"step": 360
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"entropy": 0.9640696182847023,
|
| 451 |
+
"epoch": 1.16,
|
| 452 |
+
"grad_norm": 0.36158645153045654,
|
| 453 |
+
"learning_rate": 1.5668172551691174e-07,
|
| 454 |
+
"loss": 0.9559,
|
| 455 |
+
"mean_token_accuracy": 0.7369257047772407,
|
| 456 |
+
"num_tokens": 1514434.0,
|
| 457 |
+
"step": 370
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"entropy": 0.9180433616042137,
|
| 461 |
+
"epoch": 1.1913725490196079,
|
| 462 |
+
"grad_norm": 0.33794501423835754,
|
| 463 |
+
"learning_rate": 7.210548257874927e-08,
|
| 464 |
+
"loss": 0.8909,
|
| 465 |
+
"mean_token_accuracy": 0.7486803516745567,
|
| 466 |
+
"num_tokens": 1555394.0,
|
| 467 |
+
"step": 380
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"entropy": 0.9517706975340843,
|
| 471 |
+
"epoch": 1.2227450980392156,
|
| 472 |
+
"grad_norm": 0.365311861038208,
|
| 473 |
+
"learning_rate": 1.981867244139124e-08,
|
| 474 |
+
"loss": 0.9339,
|
| 475 |
+
"mean_token_accuracy": 0.7413001000881195,
|
| 476 |
+
"num_tokens": 1596354.0,
|
| 477 |
+
"step": 390
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"entropy": 0.9845273345708847,
|
| 481 |
+
"epoch": 1.2541176470588236,
|
| 482 |
+
"grad_norm": 0.3574397563934326,
|
| 483 |
+
"learning_rate": 1.6389810421846286e-10,
|
| 484 |
+
"loss": 0.9673,
|
| 485 |
+
"mean_token_accuracy": 0.7334555193781853,
|
| 486 |
+
"num_tokens": 1637314.0,
|
| 487 |
+
"step": 400
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 1.2541176470588236,
|
| 491 |
+
"eval_entropy": 0.9607294366094801,
|
| 492 |
+
"eval_loss": 0.9414717555046082,
|
| 493 |
+
"eval_mean_token_accuracy": 0.7371565114127265,
|
| 494 |
+
"eval_num_tokens": 1637314.0,
|
| 495 |
+
"eval_runtime": 55.7023,
|
| 496 |
+
"eval_samples_per_second": 4.039,
|
| 497 |
+
"eval_steps_per_second": 4.039,
|
| 498 |
+
"step": 400
|
| 499 |
+
}
|
| 500 |
+
],
|
| 501 |
+
"logging_steps": 10,
|
| 502 |
+
"max_steps": 400,
|
| 503 |
+
"num_input_tokens_seen": 0,
|
| 504 |
+
"num_train_epochs": 2,
|
| 505 |
+
"save_steps": 100,
|
| 506 |
+
"stateful_callbacks": {
|
| 507 |
+
"TrainerControl": {
|
| 508 |
+
"args": {
|
| 509 |
+
"should_epoch_stop": false,
|
| 510 |
+
"should_evaluate": false,
|
| 511 |
+
"should_log": false,
|
| 512 |
+
"should_save": true,
|
| 513 |
+
"should_training_stop": true
|
| 514 |
+
},
|
| 515 |
+
"attributes": {}
|
| 516 |
+
}
|
| 517 |
+
},
|
| 518 |
+
"total_flos": 7.522663976342323e+16,
|
| 519 |
+
"train_batch_size": 1,
|
| 520 |
+
"trial_name": null,
|
| 521 |
+
"trial_params": null
|
| 522 |
+
}
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "</s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": "</s>",
|
| 17 |
+
"unk_token": {
|
| 18 |
+
"content": "<unk>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": false,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
}
|
| 24 |
+
}
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d7edbeaf20dd7f571b5dd1c54d9ace4f9b6299127cc7ba2afb14a6d51a4a79a4
|
| 3 |
+
size 17078136
|
tokenizer_config.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|