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 |
+
"v_proj",
|
| 29 |
+
"o_proj",
|
| 30 |
+
"gate_proj",
|
| 31 |
+
"down_proj",
|
| 32 |
+
"up_proj",
|
| 33 |
+
"k_proj",
|
| 34 |
+
"q_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": 0.6829276084899902,
|
| 4 |
+
"best_model_checkpoint": "./python-code-explaining-ministral-8B/checkpoint-200",
|
| 5 |
+
"epoch": 0.9411764705882353,
|
| 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": 1.4653201967477798,
|
| 14 |
+
"epoch": 0.047058823529411764,
|
| 15 |
+
"grad_norm": 0.7277931571006775,
|
| 16 |
+
"learning_rate": 7.500000000000001e-06,
|
| 17 |
+
"loss": 2.0127,
|
| 18 |
+
"mean_token_accuracy": 0.5873390108346939,
|
| 19 |
+
"num_tokens": 15889.0,
|
| 20 |
+
"step": 10
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"entropy": 1.4459994971752166,
|
| 24 |
+
"epoch": 0.09411764705882353,
|
| 25 |
+
"grad_norm": 0.5549399256706238,
|
| 26 |
+
"learning_rate": 9.991971098713136e-06,
|
| 27 |
+
"loss": 1.7954,
|
| 28 |
+
"mean_token_accuracy": 0.6094311743974685,
|
| 29 |
+
"num_tokens": 35503.0,
|
| 30 |
+
"step": 20
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"entropy": 1.5394100606441499,
|
| 34 |
+
"epoch": 0.1411764705882353,
|
| 35 |
+
"grad_norm": 0.6845958232879639,
|
| 36 |
+
"learning_rate": 9.952707929045018e-06,
|
| 37 |
+
"loss": 1.7686,
|
| 38 |
+
"mean_token_accuracy": 0.6184274405241013,
|
| 39 |
+
"num_tokens": 51214.0,
|
| 40 |
+
"step": 30
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"entropy": 1.4925685793161392,
|
| 44 |
+
"epoch": 0.18823529411764706,
|
| 45 |
+
"grad_norm": 0.5904952883720398,
|
| 46 |
+
"learning_rate": 9.88099273969759e-06,
|
| 47 |
+
"loss": 1.4545,
|
| 48 |
+
"mean_token_accuracy": 0.6649785250425339,
|
| 49 |
+
"num_tokens": 68991.0,
|
| 50 |
+
"step": 40
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"entropy": 1.2551275834441185,
|
| 54 |
+
"epoch": 0.23529411764705882,
|
| 55 |
+
"grad_norm": 0.5888926982879639,
|
| 56 |
+
"learning_rate": 9.777295435775165e-06,
|
| 57 |
+
"loss": 1.2224,
|
| 58 |
+
"mean_token_accuracy": 0.7385784611105919,
|
| 59 |
+
"num_tokens": 87871.0,
|
| 60 |
+
"step": 50
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 0.23529411764705882,
|
| 64 |
+
"eval_entropy": 1.0665295275052389,
|
| 65 |
+
"eval_loss": 1.0585410594940186,
|
| 66 |
+
"eval_mean_token_accuracy": 0.7891709744930268,
|
| 67 |
+
"eval_num_tokens": 87871.0,
|
| 68 |
+
"eval_runtime": 19.7003,
|
| 69 |
+
"eval_samples_per_second": 7.614,
|
| 70 |
+
"eval_steps_per_second": 7.614,
|
| 71 |
+
"step": 50
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"entropy": 0.9183753415942192,
|
| 75 |
+
"epoch": 0.2823529411764706,
|
| 76 |
+
"grad_norm": 0.4278884828090668,
|
| 77 |
+
"learning_rate": 9.64229548131959e-06,
|
| 78 |
+
"loss": 0.9356,
|
| 79 |
+
"mean_token_accuracy": 0.8226072296500206,
|
| 80 |
+
"num_tokens": 103270.0,
|
| 81 |
+
"step": 60
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"entropy": 0.8008665464818477,
|
| 85 |
+
"epoch": 0.32941176470588235,
|
| 86 |
+
"grad_norm": 0.45098593831062317,
|
| 87 |
+
"learning_rate": 9.476877447204308e-06,
|
| 88 |
+
"loss": 0.8929,
|
| 89 |
+
"mean_token_accuracy": 0.8326325386762619,
|
| 90 |
+
"num_tokens": 120480.0,
|
| 91 |
+
"step": 70
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"entropy": 0.7590022340416909,
|
| 95 |
+
"epoch": 0.3764705882352941,
|
| 96 |
+
"grad_norm": 0.33732372522354126,
|
| 97 |
+
"learning_rate": 9.282125215090694e-06,
|
| 98 |
+
"loss": 0.8101,
|
| 99 |
+
"mean_token_accuracy": 0.8484514832496644,
|
| 100 |
+
"num_tokens": 138964.0,
|
| 101 |
+
"step": 80
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"entropy": 0.7300104968249798,
|
| 105 |
+
"epoch": 0.4235294117647059,
|
| 106 |
+
"grad_norm": 0.28080910444259644,
|
| 107 |
+
"learning_rate": 9.059314875424553e-06,
|
| 108 |
+
"loss": 0.748,
|
| 109 |
+
"mean_token_accuracy": 0.8571715533733368,
|
| 110 |
+
"num_tokens": 156022.0,
|
| 111 |
+
"step": 90
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"entropy": 0.8075251087546349,
|
| 115 |
+
"epoch": 0.47058823529411764,
|
| 116 |
+
"grad_norm": 0.2512107193470001,
|
| 117 |
+
"learning_rate": 8.809906366007832e-06,
|
| 118 |
+
"loss": 0.8393,
|
| 119 |
+
"mean_token_accuracy": 0.8405888199806213,
|
| 120 |
+
"num_tokens": 175876.0,
|
| 121 |
+
"step": 100
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.47058823529411764,
|
| 125 |
+
"eval_entropy": 0.741940244436264,
|
| 126 |
+
"eval_loss": 0.7406004071235657,
|
| 127 |
+
"eval_mean_token_accuracy": 0.8551796130339304,
|
| 128 |
+
"eval_num_tokens": 175876.0,
|
| 129 |
+
"eval_runtime": 19.7066,
|
| 130 |
+
"eval_samples_per_second": 7.612,
|
| 131 |
+
"eval_steps_per_second": 7.612,
|
| 132 |
+
"step": 100
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"entropy": 0.7609439708292485,
|
| 136 |
+
"epoch": 0.5176470588235295,
|
| 137 |
+
"grad_norm": 0.2839739918708801,
|
| 138 |
+
"learning_rate": 8.535533905932739e-06,
|
| 139 |
+
"loss": 0.7876,
|
| 140 |
+
"mean_token_accuracy": 0.8530501529574395,
|
| 141 |
+
"num_tokens": 193834.0,
|
| 142 |
+
"step": 110
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"entropy": 0.7604947410523891,
|
| 146 |
+
"epoch": 0.5647058823529412,
|
| 147 |
+
"grad_norm": 0.2849130928516388,
|
| 148 |
+
"learning_rate": 8.237995287558801e-06,
|
| 149 |
+
"loss": 0.7681,
|
| 150 |
+
"mean_token_accuracy": 0.8534491807222366,
|
| 151 |
+
"num_tokens": 212048.0,
|
| 152 |
+
"step": 120
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"entropy": 0.7774970278143882,
|
| 156 |
+
"epoch": 0.611764705882353,
|
| 157 |
+
"grad_norm": 0.2741145193576813,
|
| 158 |
+
"learning_rate": 7.919240096695877e-06,
|
| 159 |
+
"loss": 0.7804,
|
| 160 |
+
"mean_token_accuracy": 0.846380078792572,
|
| 161 |
+
"num_tokens": 231794.0,
|
| 162 |
+
"step": 130
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"entropy": 0.7487856037914753,
|
| 166 |
+
"epoch": 0.6588235294117647,
|
| 167 |
+
"grad_norm": 0.2995753884315491,
|
| 168 |
+
"learning_rate": 7.58135693817893e-06,
|
| 169 |
+
"loss": 0.7511,
|
| 170 |
+
"mean_token_accuracy": 0.855742147564888,
|
| 171 |
+
"num_tokens": 251107.0,
|
| 172 |
+
"step": 140
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"entropy": 0.7198293171823025,
|
| 176 |
+
"epoch": 0.7058823529411765,
|
| 177 |
+
"grad_norm": 0.2929564416408539,
|
| 178 |
+
"learning_rate": 7.226559750537462e-06,
|
| 179 |
+
"loss": 0.7237,
|
| 180 |
+
"mean_token_accuracy": 0.8601349666714668,
|
| 181 |
+
"num_tokens": 268729.0,
|
| 182 |
+
"step": 150
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.7058823529411765,
|
| 186 |
+
"eval_entropy": 0.7403583451112111,
|
| 187 |
+
"eval_loss": 0.698711633682251,
|
| 188 |
+
"eval_mean_token_accuracy": 0.8560926202932994,
|
| 189 |
+
"eval_num_tokens": 268729.0,
|
| 190 |
+
"eval_runtime": 19.698,
|
| 191 |
+
"eval_samples_per_second": 7.615,
|
| 192 |
+
"eval_steps_per_second": 7.615,
|
| 193 |
+
"step": 150
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"entropy": 0.7665069714188576,
|
| 197 |
+
"epoch": 0.7529411764705882,
|
| 198 |
+
"grad_norm": 0.24647395312786102,
|
| 199 |
+
"learning_rate": 6.857173299431084e-06,
|
| 200 |
+
"loss": 0.7298,
|
| 201 |
+
"mean_token_accuracy": 0.8562039688229561,
|
| 202 |
+
"num_tokens": 286777.0,
|
| 203 |
+
"step": 160
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"entropy": 0.6268041007220745,
|
| 207 |
+
"epoch": 0.8,
|
| 208 |
+
"grad_norm": 0.22048509120941162,
|
| 209 |
+
"learning_rate": 6.475617944903693e-06,
|
| 210 |
+
"loss": 0.6185,
|
| 211 |
+
"mean_token_accuracy": 0.8762901470065116,
|
| 212 |
+
"num_tokens": 304268.0,
|
| 213 |
+
"step": 170
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"entropy": 0.7007844015955925,
|
| 217 |
+
"epoch": 0.8470588235294118,
|
| 218 |
+
"grad_norm": 0.2350086122751236,
|
| 219 |
+
"learning_rate": 6.084393782267039e-06,
|
| 220 |
+
"loss": 0.6875,
|
| 221 |
+
"mean_token_accuracy": 0.8667616620659828,
|
| 222 |
+
"num_tokens": 321505.0,
|
| 223 |
+
"step": 180
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"entropy": 0.7066945165395737,
|
| 227 |
+
"epoch": 0.8941176470588236,
|
| 228 |
+
"grad_norm": 0.27258384227752686,
|
| 229 |
+
"learning_rate": 5.686064260528577e-06,
|
| 230 |
+
"loss": 0.7017,
|
| 231 |
+
"mean_token_accuracy": 0.8596053853631019,
|
| 232 |
+
"num_tokens": 340285.0,
|
| 233 |
+
"step": 190
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"entropy": 0.7212481118738652,
|
| 237 |
+
"epoch": 0.9411764705882353,
|
| 238 |
+
"grad_norm": 0.25644779205322266,
|
| 239 |
+
"learning_rate": 5.283239385701882e-06,
|
| 240 |
+
"loss": 0.715,
|
| 241 |
+
"mean_token_accuracy": 0.8563221275806427,
|
| 242 |
+
"num_tokens": 360365.0,
|
| 243 |
+
"step": 200
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 0.9411764705882353,
|
| 247 |
+
"eval_entropy": 0.7153916182120641,
|
| 248 |
+
"eval_loss": 0.6829276084899902,
|
| 249 |
+
"eval_mean_token_accuracy": 0.8566614667574565,
|
| 250 |
+
"eval_num_tokens": 360365.0,
|
| 251 |
+
"eval_runtime": 19.6569,
|
| 252 |
+
"eval_samples_per_second": 7.631,
|
| 253 |
+
"eval_steps_per_second": 7.631,
|
| 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": 1.655702451597312e+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 |
+
"v_proj",
|
| 29 |
+
"o_proj",
|
| 30 |
+
"gate_proj",
|
| 31 |
+
"down_proj",
|
| 32 |
+
"up_proj",
|
| 33 |
+
"k_proj",
|
| 34 |
+
"q_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": 0.675133466720581,
|
| 4 |
+
"best_model_checkpoint": "./python-code-explaining-ministral-8B/checkpoint-300",
|
| 5 |
+
"epoch": 1.4094117647058824,
|
| 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": 1.4653201967477798,
|
| 14 |
+
"epoch": 0.047058823529411764,
|
| 15 |
+
"grad_norm": 0.7277931571006775,
|
| 16 |
+
"learning_rate": 7.500000000000001e-06,
|
| 17 |
+
"loss": 2.0127,
|
| 18 |
+
"mean_token_accuracy": 0.5873390108346939,
|
| 19 |
+
"num_tokens": 15889.0,
|
| 20 |
+
"step": 10
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"entropy": 1.4459994971752166,
|
| 24 |
+
"epoch": 0.09411764705882353,
|
| 25 |
+
"grad_norm": 0.5549399256706238,
|
| 26 |
+
"learning_rate": 9.991971098713136e-06,
|
| 27 |
+
"loss": 1.7954,
|
| 28 |
+
"mean_token_accuracy": 0.6094311743974685,
|
| 29 |
+
"num_tokens": 35503.0,
|
| 30 |
+
"step": 20
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"entropy": 1.5394100606441499,
|
| 34 |
+
"epoch": 0.1411764705882353,
|
| 35 |
+
"grad_norm": 0.6845958232879639,
|
| 36 |
+
"learning_rate": 9.952707929045018e-06,
|
| 37 |
+
"loss": 1.7686,
|
| 38 |
+
"mean_token_accuracy": 0.6184274405241013,
|
| 39 |
+
"num_tokens": 51214.0,
|
| 40 |
+
"step": 30
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"entropy": 1.4925685793161392,
|
| 44 |
+
"epoch": 0.18823529411764706,
|
| 45 |
+
"grad_norm": 0.5904952883720398,
|
| 46 |
+
"learning_rate": 9.88099273969759e-06,
|
| 47 |
+
"loss": 1.4545,
|
| 48 |
+
"mean_token_accuracy": 0.6649785250425339,
|
| 49 |
+
"num_tokens": 68991.0,
|
| 50 |
+
"step": 40
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"entropy": 1.2551275834441185,
|
| 54 |
+
"epoch": 0.23529411764705882,
|
| 55 |
+
"grad_norm": 0.5888926982879639,
|
| 56 |
+
"learning_rate": 9.777295435775165e-06,
|
| 57 |
+
"loss": 1.2224,
|
| 58 |
+
"mean_token_accuracy": 0.7385784611105919,
|
| 59 |
+
"num_tokens": 87871.0,
|
| 60 |
+
"step": 50
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 0.23529411764705882,
|
| 64 |
+
"eval_entropy": 1.0665295275052389,
|
| 65 |
+
"eval_loss": 1.0585410594940186,
|
| 66 |
+
"eval_mean_token_accuracy": 0.7891709744930268,
|
| 67 |
+
"eval_num_tokens": 87871.0,
|
| 68 |
+
"eval_runtime": 19.7003,
|
| 69 |
+
"eval_samples_per_second": 7.614,
|
| 70 |
+
"eval_steps_per_second": 7.614,
|
| 71 |
+
"step": 50
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"entropy": 0.9183753415942192,
|
| 75 |
+
"epoch": 0.2823529411764706,
|
| 76 |
+
"grad_norm": 0.4278884828090668,
|
| 77 |
+
"learning_rate": 9.64229548131959e-06,
|
| 78 |
+
"loss": 0.9356,
|
| 79 |
+
"mean_token_accuracy": 0.8226072296500206,
|
| 80 |
+
"num_tokens": 103270.0,
|
| 81 |
+
"step": 60
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"entropy": 0.8008665464818477,
|
| 85 |
+
"epoch": 0.32941176470588235,
|
| 86 |
+
"grad_norm": 0.45098593831062317,
|
| 87 |
+
"learning_rate": 9.476877447204308e-06,
|
| 88 |
+
"loss": 0.8929,
|
| 89 |
+
"mean_token_accuracy": 0.8326325386762619,
|
| 90 |
+
"num_tokens": 120480.0,
|
| 91 |
+
"step": 70
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"entropy": 0.7590022340416909,
|
| 95 |
+
"epoch": 0.3764705882352941,
|
| 96 |
+
"grad_norm": 0.33732372522354126,
|
| 97 |
+
"learning_rate": 9.282125215090694e-06,
|
| 98 |
+
"loss": 0.8101,
|
| 99 |
+
"mean_token_accuracy": 0.8484514832496644,
|
| 100 |
+
"num_tokens": 138964.0,
|
| 101 |
+
"step": 80
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"entropy": 0.7300104968249798,
|
| 105 |
+
"epoch": 0.4235294117647059,
|
| 106 |
+
"grad_norm": 0.28080910444259644,
|
| 107 |
+
"learning_rate": 9.059314875424553e-06,
|
| 108 |
+
"loss": 0.748,
|
| 109 |
+
"mean_token_accuracy": 0.8571715533733368,
|
| 110 |
+
"num_tokens": 156022.0,
|
| 111 |
+
"step": 90
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"entropy": 0.8075251087546349,
|
| 115 |
+
"epoch": 0.47058823529411764,
|
| 116 |
+
"grad_norm": 0.2512107193470001,
|
| 117 |
+
"learning_rate": 8.809906366007832e-06,
|
| 118 |
+
"loss": 0.8393,
|
| 119 |
+
"mean_token_accuracy": 0.8405888199806213,
|
| 120 |
+
"num_tokens": 175876.0,
|
| 121 |
+
"step": 100
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.47058823529411764,
|
| 125 |
+
"eval_entropy": 0.741940244436264,
|
| 126 |
+
"eval_loss": 0.7406004071235657,
|
| 127 |
+
"eval_mean_token_accuracy": 0.8551796130339304,
|
| 128 |
+
"eval_num_tokens": 175876.0,
|
| 129 |
+
"eval_runtime": 19.7066,
|
| 130 |
+
"eval_samples_per_second": 7.612,
|
| 131 |
+
"eval_steps_per_second": 7.612,
|
| 132 |
+
"step": 100
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"entropy": 0.7609439708292485,
|
| 136 |
+
"epoch": 0.5176470588235295,
|
| 137 |
+
"grad_norm": 0.2839739918708801,
|
| 138 |
+
"learning_rate": 8.535533905932739e-06,
|
| 139 |
+
"loss": 0.7876,
|
| 140 |
+
"mean_token_accuracy": 0.8530501529574395,
|
| 141 |
+
"num_tokens": 193834.0,
|
| 142 |
+
"step": 110
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"entropy": 0.7604947410523891,
|
| 146 |
+
"epoch": 0.5647058823529412,
|
| 147 |
+
"grad_norm": 0.2849130928516388,
|
| 148 |
+
"learning_rate": 8.237995287558801e-06,
|
| 149 |
+
"loss": 0.7681,
|
| 150 |
+
"mean_token_accuracy": 0.8534491807222366,
|
| 151 |
+
"num_tokens": 212048.0,
|
| 152 |
+
"step": 120
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"entropy": 0.7774970278143882,
|
| 156 |
+
"epoch": 0.611764705882353,
|
| 157 |
+
"grad_norm": 0.2741145193576813,
|
| 158 |
+
"learning_rate": 7.919240096695877e-06,
|
| 159 |
+
"loss": 0.7804,
|
| 160 |
+
"mean_token_accuracy": 0.846380078792572,
|
| 161 |
+
"num_tokens": 231794.0,
|
| 162 |
+
"step": 130
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"entropy": 0.7487856037914753,
|
| 166 |
+
"epoch": 0.6588235294117647,
|
| 167 |
+
"grad_norm": 0.2995753884315491,
|
| 168 |
+
"learning_rate": 7.58135693817893e-06,
|
| 169 |
+
"loss": 0.7511,
|
| 170 |
+
"mean_token_accuracy": 0.855742147564888,
|
| 171 |
+
"num_tokens": 251107.0,
|
| 172 |
+
"step": 140
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"entropy": 0.7198293171823025,
|
| 176 |
+
"epoch": 0.7058823529411765,
|
| 177 |
+
"grad_norm": 0.2929564416408539,
|
| 178 |
+
"learning_rate": 7.226559750537462e-06,
|
| 179 |
+
"loss": 0.7237,
|
| 180 |
+
"mean_token_accuracy": 0.8601349666714668,
|
| 181 |
+
"num_tokens": 268729.0,
|
| 182 |
+
"step": 150
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.7058823529411765,
|
| 186 |
+
"eval_entropy": 0.7403583451112111,
|
| 187 |
+
"eval_loss": 0.698711633682251,
|
| 188 |
+
"eval_mean_token_accuracy": 0.8560926202932994,
|
| 189 |
+
"eval_num_tokens": 268729.0,
|
| 190 |
+
"eval_runtime": 19.698,
|
| 191 |
+
"eval_samples_per_second": 7.615,
|
| 192 |
+
"eval_steps_per_second": 7.615,
|
| 193 |
+
"step": 150
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"entropy": 0.7665069714188576,
|
| 197 |
+
"epoch": 0.7529411764705882,
|
| 198 |
+
"grad_norm": 0.24647395312786102,
|
| 199 |
+
"learning_rate": 6.857173299431084e-06,
|
| 200 |
+
"loss": 0.7298,
|
| 201 |
+
"mean_token_accuracy": 0.8562039688229561,
|
| 202 |
+
"num_tokens": 286777.0,
|
| 203 |
+
"step": 160
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"entropy": 0.6268041007220745,
|
| 207 |
+
"epoch": 0.8,
|
| 208 |
+
"grad_norm": 0.22048509120941162,
|
| 209 |
+
"learning_rate": 6.475617944903693e-06,
|
| 210 |
+
"loss": 0.6185,
|
| 211 |
+
"mean_token_accuracy": 0.8762901470065116,
|
| 212 |
+
"num_tokens": 304268.0,
|
| 213 |
+
"step": 170
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"entropy": 0.7007844015955925,
|
| 217 |
+
"epoch": 0.8470588235294118,
|
| 218 |
+
"grad_norm": 0.2350086122751236,
|
| 219 |
+
"learning_rate": 6.084393782267039e-06,
|
| 220 |
+
"loss": 0.6875,
|
| 221 |
+
"mean_token_accuracy": 0.8667616620659828,
|
| 222 |
+
"num_tokens": 321505.0,
|
| 223 |
+
"step": 180
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"entropy": 0.7066945165395737,
|
| 227 |
+
"epoch": 0.8941176470588236,
|
| 228 |
+
"grad_norm": 0.27258384227752686,
|
| 229 |
+
"learning_rate": 5.686064260528577e-06,
|
| 230 |
+
"loss": 0.7017,
|
| 231 |
+
"mean_token_accuracy": 0.8596053853631019,
|
| 232 |
+
"num_tokens": 340285.0,
|
| 233 |
+
"step": 190
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"entropy": 0.7212481118738652,
|
| 237 |
+
"epoch": 0.9411764705882353,
|
| 238 |
+
"grad_norm": 0.25644779205322266,
|
| 239 |
+
"learning_rate": 5.283239385701882e-06,
|
| 240 |
+
"loss": 0.715,
|
| 241 |
+
"mean_token_accuracy": 0.8563221275806427,
|
| 242 |
+
"num_tokens": 360365.0,
|
| 243 |
+
"step": 200
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 0.9411764705882353,
|
| 247 |
+
"eval_entropy": 0.7153916182120641,
|
| 248 |
+
"eval_loss": 0.6829276084899902,
|
| 249 |
+
"eval_mean_token_accuracy": 0.8566614667574565,
|
| 250 |
+
"eval_num_tokens": 360365.0,
|
| 251 |
+
"eval_runtime": 19.6569,
|
| 252 |
+
"eval_samples_per_second": 7.631,
|
| 253 |
+
"eval_steps_per_second": 7.631,
|
| 254 |
+
"step": 200
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"entropy": 0.673030859977007,
|
| 258 |
+
"epoch": 0.9882352941176471,
|
| 259 |
+
"grad_norm": 0.24406129121780396,
|
| 260 |
+
"learning_rate": 4.8785586190579355e-06,
|
| 261 |
+
"loss": 0.6335,
|
| 262 |
+
"mean_token_accuracy": 0.8676866918802262,
|
| 263 |
+
"num_tokens": 378251.0,
|
| 264 |
+
"step": 210
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"entropy": 0.6841881047738226,
|
| 268 |
+
"epoch": 1.0329411764705883,
|
| 269 |
+
"grad_norm": 0.27890220284461975,
|
| 270 |
+
"learning_rate": 4.474673582374381e-06,
|
| 271 |
+
"loss": 0.6552,
|
| 272 |
+
"mean_token_accuracy": 0.8625177725365287,
|
| 273 |
+
"num_tokens": 395247.0,
|
| 274 |
+
"step": 220
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"entropy": 0.7025647312402725,
|
| 278 |
+
"epoch": 1.08,
|
| 279 |
+
"grad_norm": 0.327068954706192,
|
| 280 |
+
"learning_rate": 4.074230683504541e-06,
|
| 281 |
+
"loss": 0.7156,
|
| 282 |
+
"mean_token_accuracy": 0.8613501340150833,
|
| 283 |
+
"num_tokens": 412454.0,
|
| 284 |
+
"step": 230
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"entropy": 0.689746516942978,
|
| 288 |
+
"epoch": 1.1270588235294117,
|
| 289 |
+
"grad_norm": 0.21584568917751312,
|
| 290 |
+
"learning_rate": 3.679853776110035e-06,
|
| 291 |
+
"loss": 0.6669,
|
| 292 |
+
"mean_token_accuracy": 0.8649640858173371,
|
| 293 |
+
"num_tokens": 429678.0,
|
| 294 |
+
"step": 240
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"entropy": 0.6695160642266273,
|
| 298 |
+
"epoch": 1.1741176470588235,
|
| 299 |
+
"grad_norm": 0.20020084083080292,
|
| 300 |
+
"learning_rate": 3.2941269671770193e-06,
|
| 301 |
+
"loss": 0.6519,
|
| 302 |
+
"mean_token_accuracy": 0.8649600625038147,
|
| 303 |
+
"num_tokens": 447165.0,
|
| 304 |
+
"step": 250
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 1.1741176470588235,
|
| 308 |
+
"eval_entropy": 0.697356831630071,
|
| 309 |
+
"eval_loss": 0.677371621131897,
|
| 310 |
+
"eval_mean_token_accuracy": 0.8567790834108988,
|
| 311 |
+
"eval_num_tokens": 447165.0,
|
| 312 |
+
"eval_runtime": 19.6723,
|
| 313 |
+
"eval_samples_per_second": 7.625,
|
| 314 |
+
"eval_steps_per_second": 7.625,
|
| 315 |
+
"step": 250
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"entropy": 0.671580795943737,
|
| 319 |
+
"epoch": 1.2211764705882353,
|
| 320 |
+
"grad_norm": 0.2921179234981537,
|
| 321 |
+
"learning_rate": 2.9195776849677037e-06,
|
| 322 |
+
"loss": 0.6595,
|
| 323 |
+
"mean_token_accuracy": 0.8643461763858795,
|
| 324 |
+
"num_tokens": 463630.0,
|
| 325 |
+
"step": 260
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"entropy": 0.7368650026619434,
|
| 329 |
+
"epoch": 1.2682352941176471,
|
| 330 |
+
"grad_norm": 0.20959194004535675,
|
| 331 |
+
"learning_rate": 2.5586601183523485e-06,
|
| 332 |
+
"loss": 0.7263,
|
| 333 |
+
"mean_token_accuracy": 0.8473540425300599,
|
| 334 |
+
"num_tokens": 485811.0,
|
| 335 |
+
"step": 270
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"entropy": 0.6865952230989933,
|
| 339 |
+
"epoch": 1.315294117647059,
|
| 340 |
+
"grad_norm": 0.2575845420360565,
|
| 341 |
+
"learning_rate": 2.213739136033533e-06,
|
| 342 |
+
"loss": 0.7001,
|
| 343 |
+
"mean_token_accuracy": 0.8603918835520744,
|
| 344 |
+
"num_tokens": 503430.0,
|
| 345 |
+
"step": 280
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"entropy": 0.7441840916872025,
|
| 349 |
+
"epoch": 1.3623529411764705,
|
| 350 |
+
"grad_norm": 0.3654911518096924,
|
| 351 |
+
"learning_rate": 1.8870747910300064e-06,
|
| 352 |
+
"loss": 0.7588,
|
| 353 |
+
"mean_token_accuracy": 0.8496843308210373,
|
| 354 |
+
"num_tokens": 521378.0,
|
| 355 |
+
"step": 290
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"entropy": 0.7090313516557216,
|
| 359 |
+
"epoch": 1.4094117647058824,
|
| 360 |
+
"grad_norm": 0.26411741971969604,
|
| 361 |
+
"learning_rate": 1.5808075119526323e-06,
|
| 362 |
+
"loss": 0.7321,
|
| 363 |
+
"mean_token_accuracy": 0.8554651826620102,
|
| 364 |
+
"num_tokens": 539718.0,
|
| 365 |
+
"step": 300
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 1.4094117647058824,
|
| 369 |
+
"eval_entropy": 0.6987444368998209,
|
| 370 |
+
"eval_loss": 0.675133466720581,
|
| 371 |
+
"eval_mean_token_accuracy": 0.8570177662372589,
|
| 372 |
+
"eval_num_tokens": 539718.0,
|
| 373 |
+
"eval_runtime": 19.6985,
|
| 374 |
+
"eval_samples_per_second": 7.615,
|
| 375 |
+
"eval_steps_per_second": 7.615,
|
| 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": 2.4797425270800384e+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 |
+
"v_proj",
|
| 29 |
+
"o_proj",
|
| 30 |
+
"gate_proj",
|
| 31 |
+
"down_proj",
|
| 32 |
+
"up_proj",
|
| 33 |
+
"k_proj",
|
| 34 |
+
"q_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": 0.674819827079773,
|
| 4 |
+
"best_model_checkpoint": "./python-code-explaining-ministral-8B/checkpoint-400",
|
| 5 |
+
"epoch": 1.88,
|
| 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": 1.4653201967477798,
|
| 14 |
+
"epoch": 0.047058823529411764,
|
| 15 |
+
"grad_norm": 0.7277931571006775,
|
| 16 |
+
"learning_rate": 7.500000000000001e-06,
|
| 17 |
+
"loss": 2.0127,
|
| 18 |
+
"mean_token_accuracy": 0.5873390108346939,
|
| 19 |
+
"num_tokens": 15889.0,
|
| 20 |
+
"step": 10
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"entropy": 1.4459994971752166,
|
| 24 |
+
"epoch": 0.09411764705882353,
|
| 25 |
+
"grad_norm": 0.5549399256706238,
|
| 26 |
+
"learning_rate": 9.991971098713136e-06,
|
| 27 |
+
"loss": 1.7954,
|
| 28 |
+
"mean_token_accuracy": 0.6094311743974685,
|
| 29 |
+
"num_tokens": 35503.0,
|
| 30 |
+
"step": 20
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"entropy": 1.5394100606441499,
|
| 34 |
+
"epoch": 0.1411764705882353,
|
| 35 |
+
"grad_norm": 0.6845958232879639,
|
| 36 |
+
"learning_rate": 9.952707929045018e-06,
|
| 37 |
+
"loss": 1.7686,
|
| 38 |
+
"mean_token_accuracy": 0.6184274405241013,
|
| 39 |
+
"num_tokens": 51214.0,
|
| 40 |
+
"step": 30
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"entropy": 1.4925685793161392,
|
| 44 |
+
"epoch": 0.18823529411764706,
|
| 45 |
+
"grad_norm": 0.5904952883720398,
|
| 46 |
+
"learning_rate": 9.88099273969759e-06,
|
| 47 |
+
"loss": 1.4545,
|
| 48 |
+
"mean_token_accuracy": 0.6649785250425339,
|
| 49 |
+
"num_tokens": 68991.0,
|
| 50 |
+
"step": 40
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"entropy": 1.2551275834441185,
|
| 54 |
+
"epoch": 0.23529411764705882,
|
| 55 |
+
"grad_norm": 0.5888926982879639,
|
| 56 |
+
"learning_rate": 9.777295435775165e-06,
|
| 57 |
+
"loss": 1.2224,
|
| 58 |
+
"mean_token_accuracy": 0.7385784611105919,
|
| 59 |
+
"num_tokens": 87871.0,
|
| 60 |
+
"step": 50
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 0.23529411764705882,
|
| 64 |
+
"eval_entropy": 1.0665295275052389,
|
| 65 |
+
"eval_loss": 1.0585410594940186,
|
| 66 |
+
"eval_mean_token_accuracy": 0.7891709744930268,
|
| 67 |
+
"eval_num_tokens": 87871.0,
|
| 68 |
+
"eval_runtime": 19.7003,
|
| 69 |
+
"eval_samples_per_second": 7.614,
|
| 70 |
+
"eval_steps_per_second": 7.614,
|
| 71 |
+
"step": 50
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"entropy": 0.9183753415942192,
|
| 75 |
+
"epoch": 0.2823529411764706,
|
| 76 |
+
"grad_norm": 0.4278884828090668,
|
| 77 |
+
"learning_rate": 9.64229548131959e-06,
|
| 78 |
+
"loss": 0.9356,
|
| 79 |
+
"mean_token_accuracy": 0.8226072296500206,
|
| 80 |
+
"num_tokens": 103270.0,
|
| 81 |
+
"step": 60
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"entropy": 0.8008665464818477,
|
| 85 |
+
"epoch": 0.32941176470588235,
|
| 86 |
+
"grad_norm": 0.45098593831062317,
|
| 87 |
+
"learning_rate": 9.476877447204308e-06,
|
| 88 |
+
"loss": 0.8929,
|
| 89 |
+
"mean_token_accuracy": 0.8326325386762619,
|
| 90 |
+
"num_tokens": 120480.0,
|
| 91 |
+
"step": 70
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"entropy": 0.7590022340416909,
|
| 95 |
+
"epoch": 0.3764705882352941,
|
| 96 |
+
"grad_norm": 0.33732372522354126,
|
| 97 |
+
"learning_rate": 9.282125215090694e-06,
|
| 98 |
+
"loss": 0.8101,
|
| 99 |
+
"mean_token_accuracy": 0.8484514832496644,
|
| 100 |
+
"num_tokens": 138964.0,
|
| 101 |
+
"step": 80
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"entropy": 0.7300104968249798,
|
| 105 |
+
"epoch": 0.4235294117647059,
|
| 106 |
+
"grad_norm": 0.28080910444259644,
|
| 107 |
+
"learning_rate": 9.059314875424553e-06,
|
| 108 |
+
"loss": 0.748,
|
| 109 |
+
"mean_token_accuracy": 0.8571715533733368,
|
| 110 |
+
"num_tokens": 156022.0,
|
| 111 |
+
"step": 90
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"entropy": 0.8075251087546349,
|
| 115 |
+
"epoch": 0.47058823529411764,
|
| 116 |
+
"grad_norm": 0.2512107193470001,
|
| 117 |
+
"learning_rate": 8.809906366007832e-06,
|
| 118 |
+
"loss": 0.8393,
|
| 119 |
+
"mean_token_accuracy": 0.8405888199806213,
|
| 120 |
+
"num_tokens": 175876.0,
|
| 121 |
+
"step": 100
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.47058823529411764,
|
| 125 |
+
"eval_entropy": 0.741940244436264,
|
| 126 |
+
"eval_loss": 0.7406004071235657,
|
| 127 |
+
"eval_mean_token_accuracy": 0.8551796130339304,
|
| 128 |
+
"eval_num_tokens": 175876.0,
|
| 129 |
+
"eval_runtime": 19.7066,
|
| 130 |
+
"eval_samples_per_second": 7.612,
|
| 131 |
+
"eval_steps_per_second": 7.612,
|
| 132 |
+
"step": 100
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"entropy": 0.7609439708292485,
|
| 136 |
+
"epoch": 0.5176470588235295,
|
| 137 |
+
"grad_norm": 0.2839739918708801,
|
| 138 |
+
"learning_rate": 8.535533905932739e-06,
|
| 139 |
+
"loss": 0.7876,
|
| 140 |
+
"mean_token_accuracy": 0.8530501529574395,
|
| 141 |
+
"num_tokens": 193834.0,
|
| 142 |
+
"step": 110
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"entropy": 0.7604947410523891,
|
| 146 |
+
"epoch": 0.5647058823529412,
|
| 147 |
+
"grad_norm": 0.2849130928516388,
|
| 148 |
+
"learning_rate": 8.237995287558801e-06,
|
| 149 |
+
"loss": 0.7681,
|
| 150 |
+
"mean_token_accuracy": 0.8534491807222366,
|
| 151 |
+
"num_tokens": 212048.0,
|
| 152 |
+
"step": 120
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"entropy": 0.7774970278143882,
|
| 156 |
+
"epoch": 0.611764705882353,
|
| 157 |
+
"grad_norm": 0.2741145193576813,
|
| 158 |
+
"learning_rate": 7.919240096695877e-06,
|
| 159 |
+
"loss": 0.7804,
|
| 160 |
+
"mean_token_accuracy": 0.846380078792572,
|
| 161 |
+
"num_tokens": 231794.0,
|
| 162 |
+
"step": 130
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"entropy": 0.7487856037914753,
|
| 166 |
+
"epoch": 0.6588235294117647,
|
| 167 |
+
"grad_norm": 0.2995753884315491,
|
| 168 |
+
"learning_rate": 7.58135693817893e-06,
|
| 169 |
+
"loss": 0.7511,
|
| 170 |
+
"mean_token_accuracy": 0.855742147564888,
|
| 171 |
+
"num_tokens": 251107.0,
|
| 172 |
+
"step": 140
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"entropy": 0.7198293171823025,
|
| 176 |
+
"epoch": 0.7058823529411765,
|
| 177 |
+
"grad_norm": 0.2929564416408539,
|
| 178 |
+
"learning_rate": 7.226559750537462e-06,
|
| 179 |
+
"loss": 0.7237,
|
| 180 |
+
"mean_token_accuracy": 0.8601349666714668,
|
| 181 |
+
"num_tokens": 268729.0,
|
| 182 |
+
"step": 150
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.7058823529411765,
|
| 186 |
+
"eval_entropy": 0.7403583451112111,
|
| 187 |
+
"eval_loss": 0.698711633682251,
|
| 188 |
+
"eval_mean_token_accuracy": 0.8560926202932994,
|
| 189 |
+
"eval_num_tokens": 268729.0,
|
| 190 |
+
"eval_runtime": 19.698,
|
| 191 |
+
"eval_samples_per_second": 7.615,
|
| 192 |
+
"eval_steps_per_second": 7.615,
|
| 193 |
+
"step": 150
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"entropy": 0.7665069714188576,
|
| 197 |
+
"epoch": 0.7529411764705882,
|
| 198 |
+
"grad_norm": 0.24647395312786102,
|
| 199 |
+
"learning_rate": 6.857173299431084e-06,
|
| 200 |
+
"loss": 0.7298,
|
| 201 |
+
"mean_token_accuracy": 0.8562039688229561,
|
| 202 |
+
"num_tokens": 286777.0,
|
| 203 |
+
"step": 160
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"entropy": 0.6268041007220745,
|
| 207 |
+
"epoch": 0.8,
|
| 208 |
+
"grad_norm": 0.22048509120941162,
|
| 209 |
+
"learning_rate": 6.475617944903693e-06,
|
| 210 |
+
"loss": 0.6185,
|
| 211 |
+
"mean_token_accuracy": 0.8762901470065116,
|
| 212 |
+
"num_tokens": 304268.0,
|
| 213 |
+
"step": 170
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"entropy": 0.7007844015955925,
|
| 217 |
+
"epoch": 0.8470588235294118,
|
| 218 |
+
"grad_norm": 0.2350086122751236,
|
| 219 |
+
"learning_rate": 6.084393782267039e-06,
|
| 220 |
+
"loss": 0.6875,
|
| 221 |
+
"mean_token_accuracy": 0.8667616620659828,
|
| 222 |
+
"num_tokens": 321505.0,
|
| 223 |
+
"step": 180
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"entropy": 0.7066945165395737,
|
| 227 |
+
"epoch": 0.8941176470588236,
|
| 228 |
+
"grad_norm": 0.27258384227752686,
|
| 229 |
+
"learning_rate": 5.686064260528577e-06,
|
| 230 |
+
"loss": 0.7017,
|
| 231 |
+
"mean_token_accuracy": 0.8596053853631019,
|
| 232 |
+
"num_tokens": 340285.0,
|
| 233 |
+
"step": 190
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"entropy": 0.7212481118738652,
|
| 237 |
+
"epoch": 0.9411764705882353,
|
| 238 |
+
"grad_norm": 0.25644779205322266,
|
| 239 |
+
"learning_rate": 5.283239385701882e-06,
|
| 240 |
+
"loss": 0.715,
|
| 241 |
+
"mean_token_accuracy": 0.8563221275806427,
|
| 242 |
+
"num_tokens": 360365.0,
|
| 243 |
+
"step": 200
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 0.9411764705882353,
|
| 247 |
+
"eval_entropy": 0.7153916182120641,
|
| 248 |
+
"eval_loss": 0.6829276084899902,
|
| 249 |
+
"eval_mean_token_accuracy": 0.8566614667574565,
|
| 250 |
+
"eval_num_tokens": 360365.0,
|
| 251 |
+
"eval_runtime": 19.6569,
|
| 252 |
+
"eval_samples_per_second": 7.631,
|
| 253 |
+
"eval_steps_per_second": 7.631,
|
| 254 |
+
"step": 200
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"entropy": 0.673030859977007,
|
| 258 |
+
"epoch": 0.9882352941176471,
|
| 259 |
+
"grad_norm": 0.24406129121780396,
|
| 260 |
+
"learning_rate": 4.8785586190579355e-06,
|
| 261 |
+
"loss": 0.6335,
|
| 262 |
+
"mean_token_accuracy": 0.8676866918802262,
|
| 263 |
+
"num_tokens": 378251.0,
|
| 264 |
+
"step": 210
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"entropy": 0.6841881047738226,
|
| 268 |
+
"epoch": 1.0329411764705883,
|
| 269 |
+
"grad_norm": 0.27890220284461975,
|
| 270 |
+
"learning_rate": 4.474673582374381e-06,
|
| 271 |
+
"loss": 0.6552,
|
| 272 |
+
"mean_token_accuracy": 0.8625177725365287,
|
| 273 |
+
"num_tokens": 395247.0,
|
| 274 |
+
"step": 220
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"entropy": 0.7025647312402725,
|
| 278 |
+
"epoch": 1.08,
|
| 279 |
+
"grad_norm": 0.327068954706192,
|
| 280 |
+
"learning_rate": 4.074230683504541e-06,
|
| 281 |
+
"loss": 0.7156,
|
| 282 |
+
"mean_token_accuracy": 0.8613501340150833,
|
| 283 |
+
"num_tokens": 412454.0,
|
| 284 |
+
"step": 230
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"entropy": 0.689746516942978,
|
| 288 |
+
"epoch": 1.1270588235294117,
|
| 289 |
+
"grad_norm": 0.21584568917751312,
|
| 290 |
+
"learning_rate": 3.679853776110035e-06,
|
| 291 |
+
"loss": 0.6669,
|
| 292 |
+
"mean_token_accuracy": 0.8649640858173371,
|
| 293 |
+
"num_tokens": 429678.0,
|
| 294 |
+
"step": 240
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"entropy": 0.6695160642266273,
|
| 298 |
+
"epoch": 1.1741176470588235,
|
| 299 |
+
"grad_norm": 0.20020084083080292,
|
| 300 |
+
"learning_rate": 3.2941269671770193e-06,
|
| 301 |
+
"loss": 0.6519,
|
| 302 |
+
"mean_token_accuracy": 0.8649600625038147,
|
| 303 |
+
"num_tokens": 447165.0,
|
| 304 |
+
"step": 250
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 1.1741176470588235,
|
| 308 |
+
"eval_entropy": 0.697356831630071,
|
| 309 |
+
"eval_loss": 0.677371621131897,
|
| 310 |
+
"eval_mean_token_accuracy": 0.8567790834108988,
|
| 311 |
+
"eval_num_tokens": 447165.0,
|
| 312 |
+
"eval_runtime": 19.6723,
|
| 313 |
+
"eval_samples_per_second": 7.625,
|
| 314 |
+
"eval_steps_per_second": 7.625,
|
| 315 |
+
"step": 250
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"entropy": 0.671580795943737,
|
| 319 |
+
"epoch": 1.2211764705882353,
|
| 320 |
+
"grad_norm": 0.2921179234981537,
|
| 321 |
+
"learning_rate": 2.9195776849677037e-06,
|
| 322 |
+
"loss": 0.6595,
|
| 323 |
+
"mean_token_accuracy": 0.8643461763858795,
|
| 324 |
+
"num_tokens": 463630.0,
|
| 325 |
+
"step": 260
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"entropy": 0.7368650026619434,
|
| 329 |
+
"epoch": 1.2682352941176471,
|
| 330 |
+
"grad_norm": 0.20959194004535675,
|
| 331 |
+
"learning_rate": 2.5586601183523485e-06,
|
| 332 |
+
"loss": 0.7263,
|
| 333 |
+
"mean_token_accuracy": 0.8473540425300599,
|
| 334 |
+
"num_tokens": 485811.0,
|
| 335 |
+
"step": 270
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"entropy": 0.6865952230989933,
|
| 339 |
+
"epoch": 1.315294117647059,
|
| 340 |
+
"grad_norm": 0.2575845420360565,
|
| 341 |
+
"learning_rate": 2.213739136033533e-06,
|
| 342 |
+
"loss": 0.7001,
|
| 343 |
+
"mean_token_accuracy": 0.8603918835520744,
|
| 344 |
+
"num_tokens": 503430.0,
|
| 345 |
+
"step": 280
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"entropy": 0.7441840916872025,
|
| 349 |
+
"epoch": 1.3623529411764705,
|
| 350 |
+
"grad_norm": 0.3654911518096924,
|
| 351 |
+
"learning_rate": 1.8870747910300064e-06,
|
| 352 |
+
"loss": 0.7588,
|
| 353 |
+
"mean_token_accuracy": 0.8496843308210373,
|
| 354 |
+
"num_tokens": 521378.0,
|
| 355 |
+
"step": 290
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"entropy": 0.7090313516557216,
|
| 359 |
+
"epoch": 1.4094117647058824,
|
| 360 |
+
"grad_norm": 0.26411741971969604,
|
| 361 |
+
"learning_rate": 1.5808075119526323e-06,
|
| 362 |
+
"loss": 0.7321,
|
| 363 |
+
"mean_token_accuracy": 0.8554651826620102,
|
| 364 |
+
"num_tokens": 539718.0,
|
| 365 |
+
"step": 300
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 1.4094117647058824,
|
| 369 |
+
"eval_entropy": 0.6987444368998209,
|
| 370 |
+
"eval_loss": 0.675133466720581,
|
| 371 |
+
"eval_mean_token_accuracy": 0.8570177662372589,
|
| 372 |
+
"eval_num_tokens": 539718.0,
|
| 373 |
+
"eval_runtime": 19.6985,
|
| 374 |
+
"eval_samples_per_second": 7.615,
|
| 375 |
+
"eval_steps_per_second": 7.615,
|
| 376 |
+
"step": 300
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"entropy": 0.6481771796941758,
|
| 380 |
+
"epoch": 1.4564705882352942,
|
| 381 |
+
"grad_norm": 0.28893569111824036,
|
| 382 |
+
"learning_rate": 1.2969440781048015e-06,
|
| 383 |
+
"loss": 0.6517,
|
| 384 |
+
"mean_token_accuracy": 0.8716490805149079,
|
| 385 |
+
"num_tokens": 556870.0,
|
| 386 |
+
"step": 310
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"entropy": 0.6920560412108898,
|
| 390 |
+
"epoch": 1.5035294117647058,
|
| 391 |
+
"grad_norm": 0.4427206814289093,
|
| 392 |
+
"learning_rate": 1.0373444703037644e-06,
|
| 393 |
+
"loss": 0.7018,
|
| 394 |
+
"mean_token_accuracy": 0.8657162368297577,
|
| 395 |
+
"num_tokens": 573830.0,
|
| 396 |
+
"step": 320
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"entropy": 0.7248241394758225,
|
| 400 |
+
"epoch": 1.5505882352941176,
|
| 401 |
+
"grad_norm": 0.2903401851654053,
|
| 402 |
+
"learning_rate": 8.037096835812885e-07,
|
| 403 |
+
"loss": 0.7149,
|
| 404 |
+
"mean_token_accuracy": 0.8487672194838524,
|
| 405 |
+
"num_tokens": 595181.0,
|
| 406 |
+
"step": 330
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"entropy": 0.6708925925195217,
|
| 410 |
+
"epoch": 1.5976470588235294,
|
| 411 |
+
"grad_norm": 0.26894035935401917,
|
| 412 |
+
"learning_rate": 5.97570581619446e-07,
|
| 413 |
+
"loss": 0.6893,
|
| 414 |
+
"mean_token_accuracy": 0.8681129783391952,
|
| 415 |
+
"num_tokens": 612826.0,
|
| 416 |
+
"step": 340
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"entropy": 0.6908136330544948,
|
| 420 |
+
"epoch": 1.644705882352941,
|
| 421 |
+
"grad_norm": 0.30128714442253113,
|
| 422 |
+
"learning_rate": 4.2027786595149555e-07,
|
| 423 |
+
"loss": 0.6891,
|
| 424 |
+
"mean_token_accuracy": 0.862765510380268,
|
| 425 |
+
"num_tokens": 630006.0,
|
| 426 |
+
"step": 350
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 1.644705882352941,
|
| 430 |
+
"eval_entropy": 0.6983365495999654,
|
| 431 |
+
"eval_loss": 0.6749418377876282,
|
| 432 |
+
"eval_mean_token_accuracy": 0.8571307412783304,
|
| 433 |
+
"eval_num_tokens": 630006.0,
|
| 434 |
+
"eval_runtime": 19.5654,
|
| 435 |
+
"eval_samples_per_second": 7.667,
|
| 436 |
+
"eval_steps_per_second": 7.667,
|
| 437 |
+
"step": 350
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"entropy": 0.6560934245586395,
|
| 441 |
+
"epoch": 1.691764705882353,
|
| 442 |
+
"grad_norm": 0.1856110543012619,
|
| 443 |
+
"learning_rate": 2.7299322565344956e-07,
|
| 444 |
+
"loss": 0.644,
|
| 445 |
+
"mean_token_accuracy": 0.8646032854914665,
|
| 446 |
+
"num_tokens": 647721.0,
|
| 447 |
+
"step": 360
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"entropy": 0.653809055685997,
|
| 451 |
+
"epoch": 1.7388235294117647,
|
| 452 |
+
"grad_norm": 0.35816699266433716,
|
| 453 |
+
"learning_rate": 1.5668172551691174e-07,
|
| 454 |
+
"loss": 0.6439,
|
| 455 |
+
"mean_token_accuracy": 0.870233541727066,
|
| 456 |
+
"num_tokens": 664737.0,
|
| 457 |
+
"step": 370
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"entropy": 0.67252416908741,
|
| 461 |
+
"epoch": 1.7858823529411765,
|
| 462 |
+
"grad_norm": 0.27327945828437805,
|
| 463 |
+
"learning_rate": 7.210548257874927e-08,
|
| 464 |
+
"loss": 0.6815,
|
| 465 |
+
"mean_token_accuracy": 0.8636096775531769,
|
| 466 |
+
"num_tokens": 683029.0,
|
| 467 |
+
"step": 380
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"entropy": 0.6893440768122673,
|
| 471 |
+
"epoch": 1.8329411764705883,
|
| 472 |
+
"grad_norm": 0.21752667427062988,
|
| 473 |
+
"learning_rate": 1.981867244139124e-08,
|
| 474 |
+
"loss": 0.6818,
|
| 475 |
+
"mean_token_accuracy": 0.8598972901701927,
|
| 476 |
+
"num_tokens": 701843.0,
|
| 477 |
+
"step": 390
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"entropy": 0.6849909484386444,
|
| 481 |
+
"epoch": 1.88,
|
| 482 |
+
"grad_norm": 0.25884613394737244,
|
| 483 |
+
"learning_rate": 1.6389810421846286e-10,
|
| 484 |
+
"loss": 0.6713,
|
| 485 |
+
"mean_token_accuracy": 0.8590405523777008,
|
| 486 |
+
"num_tokens": 720477.0,
|
| 487 |
+
"step": 400
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 1.88,
|
| 491 |
+
"eval_entropy": 0.6989880565802257,
|
| 492 |
+
"eval_loss": 0.674819827079773,
|
| 493 |
+
"eval_mean_token_accuracy": 0.8568845391273499,
|
| 494 |
+
"eval_num_tokens": 720477.0,
|
| 495 |
+
"eval_runtime": 19.6125,
|
| 496 |
+
"eval_samples_per_second": 7.648,
|
| 497 |
+
"eval_steps_per_second": 7.648,
|
| 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": 3.3102424908619776e+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
|
|
|