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 +486 -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_0.6
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# oh_teknium_scaling_down_ratiocontrolled_0.6
|
| 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: 0.5925
|
| 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_0.6
|
|
|
|
| 16 |
|
| 17 |
# oh_teknium_scaling_down_ratiocontrolled_0.6
|
| 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_0.6 dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
- Loss: 0.5925
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.9956763434218656,
|
| 3 |
+
"eval_loss": 0.5925096869468689,
|
| 4 |
+
"eval_runtime": 218.8369,
|
| 5 |
+
"eval_samples_per_second": 24.918,
|
| 6 |
+
"eval_steps_per_second": 0.393,
|
| 7 |
+
"total_flos": 1014863191080960.0,
|
| 8 |
+
"train_loss": 0.5583474431494282,
|
| 9 |
+
"train_runtime": 36039.2177,
|
| 10 |
+
"train_samples_per_second": 8.624,
|
| 11 |
+
"train_steps_per_second": 0.017
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.9956763434218656,
|
| 3 |
+
"eval_loss": 0.5925096869468689,
|
| 4 |
+
"eval_runtime": 218.8369,
|
| 5 |
+
"eval_samples_per_second": 24.918,
|
| 6 |
+
"eval_steps_per_second": 0.393
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.9956763434218656,
|
| 3 |
+
"total_flos": 1014863191080960.0,
|
| 4 |
+
"train_loss": 0.5583474431494282,
|
| 5 |
+
"train_runtime": 36039.2177,
|
| 6 |
+
"train_samples_per_second": 8.624,
|
| 7 |
+
"train_steps_per_second": 0.017
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,486 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.9956763434218656,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 606,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.04941321803582458,
|
| 13 |
+
"grad_norm": 1.6209430496465307,
|
| 14 |
+
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 0.7766,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.09882643607164916,
|
| 20 |
+
"grad_norm": 0.9519377614492518,
|
| 21 |
+
"learning_rate": 5e-06,
|
| 22 |
+
"loss": 0.701,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.14823965410747375,
|
| 27 |
+
"grad_norm": 1.2733793590976263,
|
| 28 |
+
"learning_rate": 5e-06,
|
| 29 |
+
"loss": 0.6837,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.19765287214329832,
|
| 34 |
+
"grad_norm": 3.4479149115445775,
|
| 35 |
+
"learning_rate": 5e-06,
|
| 36 |
+
"loss": 0.6695,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.24706609017912293,
|
| 41 |
+
"grad_norm": 1.2813923732794312,
|
| 42 |
+
"learning_rate": 5e-06,
|
| 43 |
+
"loss": 0.6589,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.2964793082149475,
|
| 48 |
+
"grad_norm": 0.8928779733984088,
|
| 49 |
+
"learning_rate": 5e-06,
|
| 50 |
+
"loss": 0.6449,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.3458925262507721,
|
| 55 |
+
"grad_norm": 0.8314021034907043,
|
| 56 |
+
"learning_rate": 5e-06,
|
| 57 |
+
"loss": 0.635,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.39530574428659665,
|
| 62 |
+
"grad_norm": 0.922459020394824,
|
| 63 |
+
"learning_rate": 5e-06,
|
| 64 |
+
"loss": 0.615,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.44471896232242125,
|
| 69 |
+
"grad_norm": 0.8623664802985196,
|
| 70 |
+
"learning_rate": 5e-06,
|
| 71 |
+
"loss": 0.626,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.49413218035824585,
|
| 76 |
+
"grad_norm": 0.7426980782202587,
|
| 77 |
+
"learning_rate": 5e-06,
|
| 78 |
+
"loss": 0.6203,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.5435453983940705,
|
| 83 |
+
"grad_norm": 0.6400873085432273,
|
| 84 |
+
"learning_rate": 5e-06,
|
| 85 |
+
"loss": 0.6109,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.592958616429895,
|
| 90 |
+
"grad_norm": 0.7407686872440787,
|
| 91 |
+
"learning_rate": 5e-06,
|
| 92 |
+
"loss": 0.62,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.6423718344657195,
|
| 97 |
+
"grad_norm": 0.5033937870120936,
|
| 98 |
+
"learning_rate": 5e-06,
|
| 99 |
+
"loss": 0.6073,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.6917850525015442,
|
| 104 |
+
"grad_norm": 0.5119786757364595,
|
| 105 |
+
"learning_rate": 5e-06,
|
| 106 |
+
"loss": 0.6145,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.7411982705373688,
|
| 111 |
+
"grad_norm": 0.5363822451242412,
|
| 112 |
+
"learning_rate": 5e-06,
|
| 113 |
+
"loss": 0.6064,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.7906114885731933,
|
| 118 |
+
"grad_norm": 0.44308250800006926,
|
| 119 |
+
"learning_rate": 5e-06,
|
| 120 |
+
"loss": 0.6109,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.840024706609018,
|
| 125 |
+
"grad_norm": 0.7573531381636358,
|
| 126 |
+
"learning_rate": 5e-06,
|
| 127 |
+
"loss": 0.5981,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.8894379246448425,
|
| 132 |
+
"grad_norm": 0.4708891316539144,
|
| 133 |
+
"learning_rate": 5e-06,
|
| 134 |
+
"loss": 0.6005,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.938851142680667,
|
| 139 |
+
"grad_norm": 0.5108818641260817,
|
| 140 |
+
"learning_rate": 5e-06,
|
| 141 |
+
"loss": 0.5976,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.9882643607164917,
|
| 146 |
+
"grad_norm": 0.5229851868470643,
|
| 147 |
+
"learning_rate": 5e-06,
|
| 148 |
+
"loss": 0.6028,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.9981470043236566,
|
| 153 |
+
"eval_loss": 0.6029700636863708,
|
| 154 |
+
"eval_runtime": 217.1971,
|
| 155 |
+
"eval_samples_per_second": 25.106,
|
| 156 |
+
"eval_steps_per_second": 0.396,
|
| 157 |
+
"step": 202
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 1.038295243977764,
|
| 161 |
+
"grad_norm": 0.7307392640401572,
|
| 162 |
+
"learning_rate": 5e-06,
|
| 163 |
+
"loss": 0.6118,
|
| 164 |
+
"step": 210
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 1.0877084620135886,
|
| 168 |
+
"grad_norm": 0.6306789519075718,
|
| 169 |
+
"learning_rate": 5e-06,
|
| 170 |
+
"loss": 0.5495,
|
| 171 |
+
"step": 220
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 1.1371216800494133,
|
| 175 |
+
"grad_norm": 0.551995285209052,
|
| 176 |
+
"learning_rate": 5e-06,
|
| 177 |
+
"loss": 0.5479,
|
| 178 |
+
"step": 230
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 1.1865348980852377,
|
| 182 |
+
"grad_norm": 0.5943432349215367,
|
| 183 |
+
"learning_rate": 5e-06,
|
| 184 |
+
"loss": 0.5519,
|
| 185 |
+
"step": 240
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 1.2359481161210624,
|
| 189 |
+
"grad_norm": 0.511079321999948,
|
| 190 |
+
"learning_rate": 5e-06,
|
| 191 |
+
"loss": 0.544,
|
| 192 |
+
"step": 250
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 1.285361334156887,
|
| 196 |
+
"grad_norm": 0.6893935348678021,
|
| 197 |
+
"learning_rate": 5e-06,
|
| 198 |
+
"loss": 0.5482,
|
| 199 |
+
"step": 260
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 1.3347745521927115,
|
| 203 |
+
"grad_norm": 0.47658531157162304,
|
| 204 |
+
"learning_rate": 5e-06,
|
| 205 |
+
"loss": 0.5422,
|
| 206 |
+
"step": 270
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 1.3841877702285361,
|
| 210 |
+
"grad_norm": 0.5429982587488421,
|
| 211 |
+
"learning_rate": 5e-06,
|
| 212 |
+
"loss": 0.5471,
|
| 213 |
+
"step": 280
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 1.4336009882643608,
|
| 217 |
+
"grad_norm": 0.5136339875870543,
|
| 218 |
+
"learning_rate": 5e-06,
|
| 219 |
+
"loss": 0.547,
|
| 220 |
+
"step": 290
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 1.4830142063001852,
|
| 224 |
+
"grad_norm": 0.4942539231640247,
|
| 225 |
+
"learning_rate": 5e-06,
|
| 226 |
+
"loss": 0.5437,
|
| 227 |
+
"step": 300
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 1.5324274243360099,
|
| 231 |
+
"grad_norm": 0.5619554594687349,
|
| 232 |
+
"learning_rate": 5e-06,
|
| 233 |
+
"loss": 0.5569,
|
| 234 |
+
"step": 310
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.5818406423718345,
|
| 238 |
+
"grad_norm": 0.6017870072446703,
|
| 239 |
+
"learning_rate": 5e-06,
|
| 240 |
+
"loss": 0.5493,
|
| 241 |
+
"step": 320
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.631253860407659,
|
| 245 |
+
"grad_norm": 0.6231183261084499,
|
| 246 |
+
"learning_rate": 5e-06,
|
| 247 |
+
"loss": 0.5472,
|
| 248 |
+
"step": 330
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.6806670784434836,
|
| 252 |
+
"grad_norm": 0.5305154407774024,
|
| 253 |
+
"learning_rate": 5e-06,
|
| 254 |
+
"loss": 0.5452,
|
| 255 |
+
"step": 340
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 1.7300802964793083,
|
| 259 |
+
"grad_norm": 0.4891222491855961,
|
| 260 |
+
"learning_rate": 5e-06,
|
| 261 |
+
"loss": 0.5456,
|
| 262 |
+
"step": 350
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 1.7794935145151327,
|
| 266 |
+
"grad_norm": 0.47628352818666636,
|
| 267 |
+
"learning_rate": 5e-06,
|
| 268 |
+
"loss": 0.5493,
|
| 269 |
+
"step": 360
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 1.8289067325509574,
|
| 273 |
+
"grad_norm": 0.701816473212176,
|
| 274 |
+
"learning_rate": 5e-06,
|
| 275 |
+
"loss": 0.5384,
|
| 276 |
+
"step": 370
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 1.878319950586782,
|
| 280 |
+
"grad_norm": 0.7012439346323032,
|
| 281 |
+
"learning_rate": 5e-06,
|
| 282 |
+
"loss": 0.5399,
|
| 283 |
+
"step": 380
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 1.9277331686226065,
|
| 287 |
+
"grad_norm": 0.5286704563935345,
|
| 288 |
+
"learning_rate": 5e-06,
|
| 289 |
+
"loss": 0.5411,
|
| 290 |
+
"step": 390
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 1.9771463866584311,
|
| 294 |
+
"grad_norm": 0.5673012683333805,
|
| 295 |
+
"learning_rate": 5e-06,
|
| 296 |
+
"loss": 0.5472,
|
| 297 |
+
"step": 400
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 1.996911673872761,
|
| 301 |
+
"eval_loss": 0.590600848197937,
|
| 302 |
+
"eval_runtime": 217.6985,
|
| 303 |
+
"eval_samples_per_second": 25.048,
|
| 304 |
+
"eval_steps_per_second": 0.395,
|
| 305 |
+
"step": 404
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 2.0271772699197035,
|
| 309 |
+
"grad_norm": 0.7765281517872499,
|
| 310 |
+
"learning_rate": 5e-06,
|
| 311 |
+
"loss": 0.552,
|
| 312 |
+
"step": 410
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 2.076590487955528,
|
| 316 |
+
"grad_norm": 1.0500543867702512,
|
| 317 |
+
"learning_rate": 5e-06,
|
| 318 |
+
"loss": 0.4902,
|
| 319 |
+
"step": 420
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 2.126003705991353,
|
| 323 |
+
"grad_norm": 0.7474905767239051,
|
| 324 |
+
"learning_rate": 5e-06,
|
| 325 |
+
"loss": 0.4842,
|
| 326 |
+
"step": 430
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 2.1754169240271772,
|
| 330 |
+
"grad_norm": 0.6479508769608777,
|
| 331 |
+
"learning_rate": 5e-06,
|
| 332 |
+
"loss": 0.4837,
|
| 333 |
+
"step": 440
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 2.2248301420630017,
|
| 337 |
+
"grad_norm": 0.7867258676081065,
|
| 338 |
+
"learning_rate": 5e-06,
|
| 339 |
+
"loss": 0.4909,
|
| 340 |
+
"step": 450
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 2.2742433600988265,
|
| 344 |
+
"grad_norm": 0.6120315273835369,
|
| 345 |
+
"learning_rate": 5e-06,
|
| 346 |
+
"loss": 0.4799,
|
| 347 |
+
"step": 460
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 2.323656578134651,
|
| 351 |
+
"grad_norm": 0.554601233952414,
|
| 352 |
+
"learning_rate": 5e-06,
|
| 353 |
+
"loss": 0.4955,
|
| 354 |
+
"step": 470
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 2.3730697961704754,
|
| 358 |
+
"grad_norm": 0.6726344876888688,
|
| 359 |
+
"learning_rate": 5e-06,
|
| 360 |
+
"loss": 0.4917,
|
| 361 |
+
"step": 480
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 2.4224830142063003,
|
| 365 |
+
"grad_norm": 0.5467312688188073,
|
| 366 |
+
"learning_rate": 5e-06,
|
| 367 |
+
"loss": 0.4859,
|
| 368 |
+
"step": 490
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 2.4718962322421247,
|
| 372 |
+
"grad_norm": 0.5386593127548613,
|
| 373 |
+
"learning_rate": 5e-06,
|
| 374 |
+
"loss": 0.4886,
|
| 375 |
+
"step": 500
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 2.5213094502779496,
|
| 379 |
+
"grad_norm": 0.5160802210584401,
|
| 380 |
+
"learning_rate": 5e-06,
|
| 381 |
+
"loss": 0.4896,
|
| 382 |
+
"step": 510
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 2.570722668313774,
|
| 386 |
+
"grad_norm": 0.5883400693424513,
|
| 387 |
+
"learning_rate": 5e-06,
|
| 388 |
+
"loss": 0.4897,
|
| 389 |
+
"step": 520
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 2.6201358863495985,
|
| 393 |
+
"grad_norm": 0.5237320973970665,
|
| 394 |
+
"learning_rate": 5e-06,
|
| 395 |
+
"loss": 0.4908,
|
| 396 |
+
"step": 530
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 2.669549104385423,
|
| 400 |
+
"grad_norm": 0.6263776920795008,
|
| 401 |
+
"learning_rate": 5e-06,
|
| 402 |
+
"loss": 0.4903,
|
| 403 |
+
"step": 540
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 2.718962322421248,
|
| 407 |
+
"grad_norm": 0.5400072524355068,
|
| 408 |
+
"learning_rate": 5e-06,
|
| 409 |
+
"loss": 0.4932,
|
| 410 |
+
"step": 550
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 2.7683755404570722,
|
| 414 |
+
"grad_norm": 0.5269571101533504,
|
| 415 |
+
"learning_rate": 5e-06,
|
| 416 |
+
"loss": 0.4856,
|
| 417 |
+
"step": 560
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 2.817788758492897,
|
| 421 |
+
"grad_norm": 0.7217468502789398,
|
| 422 |
+
"learning_rate": 5e-06,
|
| 423 |
+
"loss": 0.4937,
|
| 424 |
+
"step": 570
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 2.8672019765287216,
|
| 428 |
+
"grad_norm": 0.8254610188302234,
|
| 429 |
+
"learning_rate": 5e-06,
|
| 430 |
+
"loss": 0.4879,
|
| 431 |
+
"step": 580
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 2.916615194564546,
|
| 435 |
+
"grad_norm": 0.706816401411115,
|
| 436 |
+
"learning_rate": 5e-06,
|
| 437 |
+
"loss": 0.4876,
|
| 438 |
+
"step": 590
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 2.9660284126003704,
|
| 442 |
+
"grad_norm": 0.5856728257014393,
|
| 443 |
+
"learning_rate": 5e-06,
|
| 444 |
+
"loss": 0.4899,
|
| 445 |
+
"step": 600
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 2.9956763434218656,
|
| 449 |
+
"eval_loss": 0.5925096869468689,
|
| 450 |
+
"eval_runtime": 216.2532,
|
| 451 |
+
"eval_samples_per_second": 25.216,
|
| 452 |
+
"eval_steps_per_second": 0.398,
|
| 453 |
+
"step": 606
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 2.9956763434218656,
|
| 457 |
+
"step": 606,
|
| 458 |
+
"total_flos": 1014863191080960.0,
|
| 459 |
+
"train_loss": 0.5583474431494282,
|
| 460 |
+
"train_runtime": 36039.2177,
|
| 461 |
+
"train_samples_per_second": 8.624,
|
| 462 |
+
"train_steps_per_second": 0.017
|
| 463 |
+
}
|
| 464 |
+
],
|
| 465 |
+
"logging_steps": 10,
|
| 466 |
+
"max_steps": 606,
|
| 467 |
+
"num_input_tokens_seen": 0,
|
| 468 |
+
"num_train_epochs": 3,
|
| 469 |
+
"save_steps": 500,
|
| 470 |
+
"stateful_callbacks": {
|
| 471 |
+
"TrainerControl": {
|
| 472 |
+
"args": {
|
| 473 |
+
"should_epoch_stop": false,
|
| 474 |
+
"should_evaluate": false,
|
| 475 |
+
"should_log": false,
|
| 476 |
+
"should_save": true,
|
| 477 |
+
"should_training_stop": true
|
| 478 |
+
},
|
| 479 |
+
"attributes": {}
|
| 480 |
+
}
|
| 481 |
+
},
|
| 482 |
+
"total_flos": 1014863191080960.0,
|
| 483 |
+
"train_batch_size": 8,
|
| 484 |
+
"trial_name": null,
|
| 485 |
+
"trial_params": null
|
| 486 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|