End of training
Browse files- README.md +2 -1
- all_results.json +12 -0
- eval_results.json +7 -0
- train_results.json +8 -0
- trainer_state.json +416 -0
- training_eval_loss.png +0 -0
- training_loss.png +0 -0
README.md
CHANGED
|
@@ -4,6 +4,7 @@ license: llama3.1
|
|
| 4 |
base_model: meta-llama/Meta-Llama-3.1-8B
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
|
|
|
| 7 |
- generated_from_trainer
|
| 8 |
model-index:
|
| 9 |
- name: oh_teknium_scaling_down_ratiocontrolled_0.5
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# oh_teknium_scaling_down_ratiocontrolled_0.5
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
- Loss: 0.6025
|
| 21 |
|
|
|
|
| 4 |
base_model: meta-llama/Meta-Llama-3.1-8B
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
| 7 |
+
- full
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: oh_teknium_scaling_down_ratiocontrolled_0.5
|
|
|
|
| 16 |
|
| 17 |
# oh_teknium_scaling_down_ratiocontrolled_0.5
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on the mlfoundations-dev/oh_teknium_scaling_down_ratiocontrolled_0.5 dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
- Loss: 0.6025
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.9918458117123796,
|
| 3 |
+
"eval_loss": 0.6024672985076904,
|
| 4 |
+
"eval_runtime": 178.9685,
|
| 5 |
+
"eval_samples_per_second": 25.384,
|
| 6 |
+
"eval_steps_per_second": 0.397,
|
| 7 |
+
"total_flos": 844009392046080.0,
|
| 8 |
+
"train_loss": 0.5635922295706612,
|
| 9 |
+
"train_runtime": 29809.8263,
|
| 10 |
+
"train_samples_per_second": 8.685,
|
| 11 |
+
"train_steps_per_second": 0.017
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.9918458117123796,
|
| 3 |
+
"eval_loss": 0.6024672985076904,
|
| 4 |
+
"eval_runtime": 178.9685,
|
| 5 |
+
"eval_samples_per_second": 25.384,
|
| 6 |
+
"eval_steps_per_second": 0.397
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.9918458117123796,
|
| 3 |
+
"total_flos": 844009392046080.0,
|
| 4 |
+
"train_loss": 0.5635922295706612,
|
| 5 |
+
"train_runtime": 29809.8263,
|
| 6 |
+
"train_samples_per_second": 8.685,
|
| 7 |
+
"train_steps_per_second": 0.017
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,416 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.9918458117123796,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 504,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.05930318754633061,
|
| 13 |
+
"grad_norm": 1.3802219854574984,
|
| 14 |
+
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 0.7739,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.11860637509266123,
|
| 20 |
+
"grad_norm": 2.733409062302713,
|
| 21 |
+
"learning_rate": 5e-06,
|
| 22 |
+
"loss": 0.7072,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.17790956263899185,
|
| 27 |
+
"grad_norm": 1.3622548058844581,
|
| 28 |
+
"learning_rate": 5e-06,
|
| 29 |
+
"loss": 0.6883,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.23721275018532245,
|
| 34 |
+
"grad_norm": 2.1397210299903042,
|
| 35 |
+
"learning_rate": 5e-06,
|
| 36 |
+
"loss": 0.6653,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.2965159377316531,
|
| 41 |
+
"grad_norm": 1.1868267718380074,
|
| 42 |
+
"learning_rate": 5e-06,
|
| 43 |
+
"loss": 0.6578,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.3558191252779837,
|
| 48 |
+
"grad_norm": 0.768661599042398,
|
| 49 |
+
"learning_rate": 5e-06,
|
| 50 |
+
"loss": 0.6434,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.41512231282431433,
|
| 55 |
+
"grad_norm": 0.929517541921154,
|
| 56 |
+
"learning_rate": 5e-06,
|
| 57 |
+
"loss": 0.6412,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.4744255003706449,
|
| 62 |
+
"grad_norm": 1.0083217759359295,
|
| 63 |
+
"learning_rate": 5e-06,
|
| 64 |
+
"loss": 0.6305,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.5337286879169756,
|
| 69 |
+
"grad_norm": 0.7019920027719391,
|
| 70 |
+
"learning_rate": 5e-06,
|
| 71 |
+
"loss": 0.625,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.5930318754633062,
|
| 76 |
+
"grad_norm": 0.8179786383641104,
|
| 77 |
+
"learning_rate": 5e-06,
|
| 78 |
+
"loss": 0.6186,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.6523350630096367,
|
| 83 |
+
"grad_norm": 1.505043114613733,
|
| 84 |
+
"learning_rate": 5e-06,
|
| 85 |
+
"loss": 0.6109,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.7116382505559674,
|
| 90 |
+
"grad_norm": 0.652990528799698,
|
| 91 |
+
"learning_rate": 5e-06,
|
| 92 |
+
"loss": 0.6174,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.770941438102298,
|
| 97 |
+
"grad_norm": 0.8738270604177748,
|
| 98 |
+
"learning_rate": 5e-06,
|
| 99 |
+
"loss": 0.6099,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.8302446256486287,
|
| 104 |
+
"grad_norm": 0.7482311809054135,
|
| 105 |
+
"learning_rate": 5e-06,
|
| 106 |
+
"loss": 0.6152,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.8895478131949592,
|
| 111 |
+
"grad_norm": 0.5898841945150782,
|
| 112 |
+
"learning_rate": 5e-06,
|
| 113 |
+
"loss": 0.604,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.9488510007412898,
|
| 118 |
+
"grad_norm": 0.5509255300448996,
|
| 119 |
+
"learning_rate": 5e-06,
|
| 120 |
+
"loss": 0.6088,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.9962935507783544,
|
| 125 |
+
"eval_loss": 0.6123025417327881,
|
| 126 |
+
"eval_runtime": 177.7519,
|
| 127 |
+
"eval_samples_per_second": 25.558,
|
| 128 |
+
"eval_steps_per_second": 0.399,
|
| 129 |
+
"step": 168
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 1.0096367679762788,
|
| 133 |
+
"grad_norm": 0.5962998315100678,
|
| 134 |
+
"learning_rate": 5e-06,
|
| 135 |
+
"loss": 0.6323,
|
| 136 |
+
"step": 170
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 1.0689399555226093,
|
| 140 |
+
"grad_norm": 0.6401517658215825,
|
| 141 |
+
"learning_rate": 5e-06,
|
| 142 |
+
"loss": 0.5587,
|
| 143 |
+
"step": 180
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 1.12824314306894,
|
| 147 |
+
"grad_norm": 0.5789193360471795,
|
| 148 |
+
"learning_rate": 5e-06,
|
| 149 |
+
"loss": 0.5532,
|
| 150 |
+
"step": 190
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 1.1875463306152705,
|
| 154 |
+
"grad_norm": 0.5007418392739746,
|
| 155 |
+
"learning_rate": 5e-06,
|
| 156 |
+
"loss": 0.5449,
|
| 157 |
+
"step": 200
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 1.2468495181616013,
|
| 161 |
+
"grad_norm": 0.742099379670997,
|
| 162 |
+
"learning_rate": 5e-06,
|
| 163 |
+
"loss": 0.5506,
|
| 164 |
+
"step": 210
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 1.3061527057079318,
|
| 168 |
+
"grad_norm": 0.6357145205835705,
|
| 169 |
+
"learning_rate": 5e-06,
|
| 170 |
+
"loss": 0.5518,
|
| 171 |
+
"step": 220
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 1.3654558932542624,
|
| 175 |
+
"grad_norm": 0.4706954425299361,
|
| 176 |
+
"learning_rate": 5e-06,
|
| 177 |
+
"loss": 0.543,
|
| 178 |
+
"step": 230
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 1.424759080800593,
|
| 182 |
+
"grad_norm": 0.6430769844169726,
|
| 183 |
+
"learning_rate": 5e-06,
|
| 184 |
+
"loss": 0.5563,
|
| 185 |
+
"step": 240
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 1.4840622683469236,
|
| 189 |
+
"grad_norm": 0.714837022060758,
|
| 190 |
+
"learning_rate": 5e-06,
|
| 191 |
+
"loss": 0.5517,
|
| 192 |
+
"step": 250
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 1.5433654558932544,
|
| 196 |
+
"grad_norm": 0.5623788095234329,
|
| 197 |
+
"learning_rate": 5e-06,
|
| 198 |
+
"loss": 0.5512,
|
| 199 |
+
"step": 260
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 1.602668643439585,
|
| 203 |
+
"grad_norm": 0.5481074950687667,
|
| 204 |
+
"learning_rate": 5e-06,
|
| 205 |
+
"loss": 0.55,
|
| 206 |
+
"step": 270
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 1.6619718309859155,
|
| 210 |
+
"grad_norm": 0.5539378521915824,
|
| 211 |
+
"learning_rate": 5e-06,
|
| 212 |
+
"loss": 0.5472,
|
| 213 |
+
"step": 280
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 1.721275018532246,
|
| 217 |
+
"grad_norm": 0.8039356878348214,
|
| 218 |
+
"learning_rate": 5e-06,
|
| 219 |
+
"loss": 0.5451,
|
| 220 |
+
"step": 290
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 1.7805782060785766,
|
| 224 |
+
"grad_norm": 0.7791331104006716,
|
| 225 |
+
"learning_rate": 5e-06,
|
| 226 |
+
"loss": 0.5477,
|
| 227 |
+
"step": 300
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 1.8398813936249074,
|
| 231 |
+
"grad_norm": 0.6989192286392514,
|
| 232 |
+
"learning_rate": 5e-06,
|
| 233 |
+
"loss": 0.5558,
|
| 234 |
+
"step": 310
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.899184581171238,
|
| 238 |
+
"grad_norm": 0.5069580060902782,
|
| 239 |
+
"learning_rate": 5e-06,
|
| 240 |
+
"loss": 0.5533,
|
| 241 |
+
"step": 320
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.9584877687175686,
|
| 245 |
+
"grad_norm": 0.5741926432541444,
|
| 246 |
+
"learning_rate": 5e-06,
|
| 247 |
+
"loss": 0.5487,
|
| 248 |
+
"step": 330
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.9940696812453669,
|
| 252 |
+
"eval_loss": 0.6003267765045166,
|
| 253 |
+
"eval_runtime": 177.8259,
|
| 254 |
+
"eval_samples_per_second": 25.547,
|
| 255 |
+
"eval_steps_per_second": 0.399,
|
| 256 |
+
"step": 336
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 2.0192735359525575,
|
| 260 |
+
"grad_norm": 0.9927432888051317,
|
| 261 |
+
"learning_rate": 5e-06,
|
| 262 |
+
"loss": 0.564,
|
| 263 |
+
"step": 340
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 2.078576723498888,
|
| 267 |
+
"grad_norm": 0.7990001146652178,
|
| 268 |
+
"learning_rate": 5e-06,
|
| 269 |
+
"loss": 0.4911,
|
| 270 |
+
"step": 350
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 2.1378799110452187,
|
| 274 |
+
"grad_norm": 0.556438381270437,
|
| 275 |
+
"learning_rate": 5e-06,
|
| 276 |
+
"loss": 0.4883,
|
| 277 |
+
"step": 360
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 2.1971830985915495,
|
| 281 |
+
"grad_norm": 0.6135178834215755,
|
| 282 |
+
"learning_rate": 5e-06,
|
| 283 |
+
"loss": 0.488,
|
| 284 |
+
"step": 370
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 2.25648628613788,
|
| 288 |
+
"grad_norm": 0.5081260960389352,
|
| 289 |
+
"learning_rate": 5e-06,
|
| 290 |
+
"loss": 0.493,
|
| 291 |
+
"step": 380
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 2.3157894736842106,
|
| 295 |
+
"grad_norm": 0.68213842322186,
|
| 296 |
+
"learning_rate": 5e-06,
|
| 297 |
+
"loss": 0.4814,
|
| 298 |
+
"step": 390
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 2.375092661230541,
|
| 302 |
+
"grad_norm": 0.5560842165539166,
|
| 303 |
+
"learning_rate": 5e-06,
|
| 304 |
+
"loss": 0.4947,
|
| 305 |
+
"step": 400
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 2.4343958487768718,
|
| 309 |
+
"grad_norm": 0.561149355084152,
|
| 310 |
+
"learning_rate": 5e-06,
|
| 311 |
+
"loss": 0.4916,
|
| 312 |
+
"step": 410
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 2.4936990363232026,
|
| 316 |
+
"grad_norm": 0.534652055102195,
|
| 317 |
+
"learning_rate": 5e-06,
|
| 318 |
+
"loss": 0.4956,
|
| 319 |
+
"step": 420
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 2.553002223869533,
|
| 323 |
+
"grad_norm": 0.5766622241099675,
|
| 324 |
+
"learning_rate": 5e-06,
|
| 325 |
+
"loss": 0.494,
|
| 326 |
+
"step": 430
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 2.6123054114158637,
|
| 330 |
+
"grad_norm": 0.5664640432180077,
|
| 331 |
+
"learning_rate": 5e-06,
|
| 332 |
+
"loss": 0.4973,
|
| 333 |
+
"step": 440
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 2.6716085989621945,
|
| 337 |
+
"grad_norm": 0.5141442287518144,
|
| 338 |
+
"learning_rate": 5e-06,
|
| 339 |
+
"loss": 0.4998,
|
| 340 |
+
"step": 450
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 2.730911786508525,
|
| 344 |
+
"grad_norm": 0.622575399798233,
|
| 345 |
+
"learning_rate": 5e-06,
|
| 346 |
+
"loss": 0.4891,
|
| 347 |
+
"step": 460
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 2.790214974054855,
|
| 351 |
+
"grad_norm": 0.6328620684014972,
|
| 352 |
+
"learning_rate": 5e-06,
|
| 353 |
+
"loss": 0.5005,
|
| 354 |
+
"step": 470
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 2.849518161601186,
|
| 358 |
+
"grad_norm": 0.5410072349219146,
|
| 359 |
+
"learning_rate": 5e-06,
|
| 360 |
+
"loss": 0.4894,
|
| 361 |
+
"step": 480
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 2.9088213491475168,
|
| 365 |
+
"grad_norm": 0.6725625016553842,
|
| 366 |
+
"learning_rate": 5e-06,
|
| 367 |
+
"loss": 0.4939,
|
| 368 |
+
"step": 490
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 2.968124536693847,
|
| 372 |
+
"grad_norm": 0.6468049088847707,
|
| 373 |
+
"learning_rate": 5e-06,
|
| 374 |
+
"loss": 0.497,
|
| 375 |
+
"step": 500
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 2.9918458117123796,
|
| 379 |
+
"eval_loss": 0.6024672985076904,
|
| 380 |
+
"eval_runtime": 178.6347,
|
| 381 |
+
"eval_samples_per_second": 25.432,
|
| 382 |
+
"eval_steps_per_second": 0.397,
|
| 383 |
+
"step": 504
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 2.9918458117123796,
|
| 387 |
+
"step": 504,
|
| 388 |
+
"total_flos": 844009392046080.0,
|
| 389 |
+
"train_loss": 0.5635922295706612,
|
| 390 |
+
"train_runtime": 29809.8263,
|
| 391 |
+
"train_samples_per_second": 8.685,
|
| 392 |
+
"train_steps_per_second": 0.017
|
| 393 |
+
}
|
| 394 |
+
],
|
| 395 |
+
"logging_steps": 10,
|
| 396 |
+
"max_steps": 504,
|
| 397 |
+
"num_input_tokens_seen": 0,
|
| 398 |
+
"num_train_epochs": 3,
|
| 399 |
+
"save_steps": 500,
|
| 400 |
+
"stateful_callbacks": {
|
| 401 |
+
"TrainerControl": {
|
| 402 |
+
"args": {
|
| 403 |
+
"should_epoch_stop": false,
|
| 404 |
+
"should_evaluate": false,
|
| 405 |
+
"should_log": false,
|
| 406 |
+
"should_save": true,
|
| 407 |
+
"should_training_stop": true
|
| 408 |
+
},
|
| 409 |
+
"attributes": {}
|
| 410 |
+
}
|
| 411 |
+
},
|
| 412 |
+
"total_flos": 844009392046080.0,
|
| 413 |
+
"train_batch_size": 8,
|
| 414 |
+
"trial_name": null,
|
| 415 |
+
"trial_params": null
|
| 416 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|