End of training
Browse files- README.md +2 -1
- all_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +497 -0
- training_loss.png +0 -0
README.md
CHANGED
|
@@ -4,6 +4,7 @@ license: apache-2.0
|
|
| 4 |
base_model: Qwen/Qwen2.5-7B-Instruct
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
|
|
|
| 7 |
- generated_from_trainer
|
| 8 |
model-index:
|
| 9 |
- name: 316_globalbatchsize64_lr1e5_epochs13
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# 316_globalbatchsize64_lr1e5_epochs13
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [Qwen/Qwen2.5-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-7B-Instruct) on
|
| 19 |
|
| 20 |
## Model description
|
| 21 |
|
|
|
|
| 4 |
base_model: Qwen/Qwen2.5-7B-Instruct
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
| 7 |
+
- full
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: 316_globalbatchsize64_lr1e5_epochs13
|
|
|
|
| 16 |
|
| 17 |
# 316_globalbatchsize64_lr1e5_epochs13
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [Qwen/Qwen2.5-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-7B-Instruct) on the mlfoundations-dev/openthoughts_316 dataset.
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 13.0,
|
| 3 |
+
"total_flos": 8.525099631326003e+16,
|
| 4 |
+
"train_loss": 0.472858812717291,
|
| 5 |
+
"train_runtime": 5413.2141,
|
| 6 |
+
"train_samples_per_second": 0.759,
|
| 7 |
+
"train_steps_per_second": 0.012
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 13.0,
|
| 3 |
+
"total_flos": 8.525099631326003e+16,
|
| 4 |
+
"train_loss": 0.472858812717291,
|
| 5 |
+
"train_runtime": 5413.2141,
|
| 6 |
+
"train_samples_per_second": 0.759,
|
| 7 |
+
"train_steps_per_second": 0.012
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,497 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 13.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 65,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.2,
|
| 13 |
+
"grad_norm": 5.582363636951145,
|
| 14 |
+
"learning_rate": 1.4285714285714286e-06,
|
| 15 |
+
"loss": 0.832,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.4,
|
| 20 |
+
"grad_norm": 6.007906389380084,
|
| 21 |
+
"learning_rate": 2.8571428571428573e-06,
|
| 22 |
+
"loss": 0.8909,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.6,
|
| 27 |
+
"grad_norm": 6.0264753137548634,
|
| 28 |
+
"learning_rate": 4.2857142857142855e-06,
|
| 29 |
+
"loss": 0.9083,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.8,
|
| 34 |
+
"grad_norm": 4.758320667876842,
|
| 35 |
+
"learning_rate": 5.7142857142857145e-06,
|
| 36 |
+
"loss": 0.877,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 1.0,
|
| 41 |
+
"grad_norm": 2.4076577062917965,
|
| 42 |
+
"learning_rate": 7.1428571428571436e-06,
|
| 43 |
+
"loss": 0.7876,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 1.2,
|
| 48 |
+
"grad_norm": 2.08237296256087,
|
| 49 |
+
"learning_rate": 8.571428571428571e-06,
|
| 50 |
+
"loss": 0.7694,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 1.4,
|
| 55 |
+
"grad_norm": 4.6171372058213995,
|
| 56 |
+
"learning_rate": 1e-05,
|
| 57 |
+
"loss": 0.8187,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 1.6,
|
| 62 |
+
"grad_norm": 5.009434012692013,
|
| 63 |
+
"learning_rate": 9.99266706925562e-06,
|
| 64 |
+
"loss": 0.7953,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 1.8,
|
| 69 |
+
"grad_norm": 6.9545840167778445,
|
| 70 |
+
"learning_rate": 9.970689785771798e-06,
|
| 71 |
+
"loss": 0.7978,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 2.0,
|
| 76 |
+
"grad_norm": 5.75089354879985,
|
| 77 |
+
"learning_rate": 9.934132612707631e-06,
|
| 78 |
+
"loss": 0.7635,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 2.2,
|
| 83 |
+
"grad_norm": 4.2068964184545585,
|
| 84 |
+
"learning_rate": 9.883102778550434e-06,
|
| 85 |
+
"loss": 0.7522,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 2.4,
|
| 90 |
+
"grad_norm": 2.4112952356601105,
|
| 91 |
+
"learning_rate": 9.817749962596115e-06,
|
| 92 |
+
"loss": 0.6875,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 2.6,
|
| 97 |
+
"grad_norm": 2.3091169308054056,
|
| 98 |
+
"learning_rate": 9.738265855914014e-06,
|
| 99 |
+
"loss": 0.6611,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 2.8,
|
| 104 |
+
"grad_norm": 2.034937499528585,
|
| 105 |
+
"learning_rate": 9.644883599083959e-06,
|
| 106 |
+
"loss": 0.6835,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 3.0,
|
| 111 |
+
"grad_norm": 1.6127281142448815,
|
| 112 |
+
"learning_rate": 9.537877098354787e-06,
|
| 113 |
+
"loss": 0.6588,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 3.2,
|
| 118 |
+
"grad_norm": 1.2420963045502502,
|
| 119 |
+
"learning_rate": 9.417560222230115e-06,
|
| 120 |
+
"loss": 0.5886,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 3.4,
|
| 125 |
+
"grad_norm": 1.3274360340575622,
|
| 126 |
+
"learning_rate": 9.284285880837947e-06,
|
| 127 |
+
"loss": 0.6684,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 3.6,
|
| 132 |
+
"grad_norm": 1.0211469602843797,
|
| 133 |
+
"learning_rate": 9.138444990784455e-06,
|
| 134 |
+
"loss": 0.6007,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 3.8,
|
| 139 |
+
"grad_norm": 0.9935488366512957,
|
| 140 |
+
"learning_rate": 8.98046532852822e-06,
|
| 141 |
+
"loss": 0.6003,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 4.0,
|
| 146 |
+
"grad_norm": 0.8078724366399843,
|
| 147 |
+
"learning_rate": 8.810810275638183e-06,
|
| 148 |
+
"loss": 0.5275,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 4.2,
|
| 153 |
+
"grad_norm": 0.8588787511486784,
|
| 154 |
+
"learning_rate": 8.629977459615655e-06,
|
| 155 |
+
"loss": 0.5663,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 4.4,
|
| 160 |
+
"grad_norm": 0.7538829403060613,
|
| 161 |
+
"learning_rate": 8.438497294267117e-06,
|
| 162 |
+
"loss": 0.5168,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 4.6,
|
| 167 |
+
"grad_norm": 0.7637932323734493,
|
| 168 |
+
"learning_rate": 8.23693142390914e-06,
|
| 169 |
+
"loss": 0.5235,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 4.8,
|
| 174 |
+
"grad_norm": 0.7643306856468832,
|
| 175 |
+
"learning_rate": 8.025871075968828e-06,
|
| 176 |
+
"loss": 0.5304,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 5.0,
|
| 181 |
+
"grad_norm": 0.6324846681806777,
|
| 182 |
+
"learning_rate": 7.805935326811913e-06,
|
| 183 |
+
"loss": 0.5152,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 5.2,
|
| 188 |
+
"grad_norm": 0.6902515181754407,
|
| 189 |
+
"learning_rate": 7.57776928588511e-06,
|
| 190 |
+
"loss": 0.5243,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 5.4,
|
| 195 |
+
"grad_norm": 0.6205487787202831,
|
| 196 |
+
"learning_rate": 7.342042203498952e-06,
|
| 197 |
+
"loss": 0.4908,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 5.6,
|
| 202 |
+
"grad_norm": 0.6281988958201551,
|
| 203 |
+
"learning_rate": 7.099445507801324e-06,
|
| 204 |
+
"loss": 0.4791,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 5.8,
|
| 209 |
+
"grad_norm": 0.6812591195712626,
|
| 210 |
+
"learning_rate": 6.850690776699574e-06,
|
| 211 |
+
"loss": 0.4376,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 6.0,
|
| 216 |
+
"grad_norm": 0.5849169270560564,
|
| 217 |
+
"learning_rate": 6.5965076506799e-06,
|
| 218 |
+
"loss": 0.4264,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 6.2,
|
| 223 |
+
"grad_norm": 0.6457985712072969,
|
| 224 |
+
"learning_rate": 6.337641692646106e-06,
|
| 225 |
+
"loss": 0.4455,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 6.4,
|
| 230 |
+
"grad_norm": 0.5635801829905707,
|
| 231 |
+
"learning_rate": 6.074852201055121e-06,
|
| 232 |
+
"loss": 0.4173,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 6.6,
|
| 237 |
+
"grad_norm": 0.5223246354583466,
|
| 238 |
+
"learning_rate": 5.808909982763825e-06,
|
| 239 |
+
"loss": 0.401,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 6.8,
|
| 244 |
+
"grad_norm": 0.5498217637405548,
|
| 245 |
+
"learning_rate": 5.540595092119709e-06,
|
| 246 |
+
"loss": 0.4238,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 7.0,
|
| 251 |
+
"grad_norm": 0.4720178214619839,
|
| 252 |
+
"learning_rate": 5.270694542927089e-06,
|
| 253 |
+
"loss": 0.3866,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 7.2,
|
| 258 |
+
"grad_norm": 0.47994799351007716,
|
| 259 |
+
"learning_rate": 5e-06,
|
| 260 |
+
"loss": 0.375,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 7.4,
|
| 265 |
+
"grad_norm": 0.4754551384737771,
|
| 266 |
+
"learning_rate": 4.729305457072913e-06,
|
| 267 |
+
"loss": 0.3888,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 7.6,
|
| 272 |
+
"grad_norm": 0.4812422271485711,
|
| 273 |
+
"learning_rate": 4.459404907880293e-06,
|
| 274 |
+
"loss": 0.3479,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 7.8,
|
| 279 |
+
"grad_norm": 0.45352582219328164,
|
| 280 |
+
"learning_rate": 4.191090017236177e-06,
|
| 281 |
+
"loss": 0.3618,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 8.0,
|
| 286 |
+
"grad_norm": 0.4065323940483181,
|
| 287 |
+
"learning_rate": 3.92514779894488e-06,
|
| 288 |
+
"loss": 0.36,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 8.2,
|
| 293 |
+
"grad_norm": 0.4227442111084603,
|
| 294 |
+
"learning_rate": 3.662358307353897e-06,
|
| 295 |
+
"loss": 0.3046,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 8.4,
|
| 300 |
+
"grad_norm": 0.42414683307987505,
|
| 301 |
+
"learning_rate": 3.403492349320101e-06,
|
| 302 |
+
"loss": 0.3519,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 8.6,
|
| 307 |
+
"grad_norm": 0.40450567963312506,
|
| 308 |
+
"learning_rate": 3.149309223300428e-06,
|
| 309 |
+
"loss": 0.3273,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 8.8,
|
| 314 |
+
"grad_norm": 0.40272719775595717,
|
| 315 |
+
"learning_rate": 2.9005544921986774e-06,
|
| 316 |
+
"loss": 0.3316,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 9.0,
|
| 321 |
+
"grad_norm": 0.41851005929072993,
|
| 322 |
+
"learning_rate": 2.65795779650105e-06,
|
| 323 |
+
"loss": 0.3278,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 9.2,
|
| 328 |
+
"grad_norm": 0.4311448019356393,
|
| 329 |
+
"learning_rate": 2.422230714114891e-06,
|
| 330 |
+
"loss": 0.3086,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 9.4,
|
| 335 |
+
"grad_norm": 0.35795131810925795,
|
| 336 |
+
"learning_rate": 2.1940646731880887e-06,
|
| 337 |
+
"loss": 0.2882,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 9.6,
|
| 342 |
+
"grad_norm": 0.386946992231837,
|
| 343 |
+
"learning_rate": 1.9741289240311757e-06,
|
| 344 |
+
"loss": 0.2919,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 9.8,
|
| 349 |
+
"grad_norm": 0.44062701543471006,
|
| 350 |
+
"learning_rate": 1.7630685760908623e-06,
|
| 351 |
+
"loss": 0.2985,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 10.0,
|
| 356 |
+
"grad_norm": 0.4144356041601527,
|
| 357 |
+
"learning_rate": 1.561502705732883e-06,
|
| 358 |
+
"loss": 0.2974,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 10.2,
|
| 363 |
+
"grad_norm": 0.3823205368468367,
|
| 364 |
+
"learning_rate": 1.370022540384347e-06,
|
| 365 |
+
"loss": 0.272,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 10.4,
|
| 370 |
+
"grad_norm": 0.40310131237975705,
|
| 371 |
+
"learning_rate": 1.1891897243618184e-06,
|
| 372 |
+
"loss": 0.296,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 10.6,
|
| 377 |
+
"grad_norm": 0.3676392905535333,
|
| 378 |
+
"learning_rate": 1.0195346714717813e-06,
|
| 379 |
+
"loss": 0.2903,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 10.8,
|
| 384 |
+
"grad_norm": 0.36362349004803485,
|
| 385 |
+
"learning_rate": 8.615550092155478e-07,
|
| 386 |
+
"loss": 0.2596,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 11.0,
|
| 391 |
+
"grad_norm": 0.3644836984135043,
|
| 392 |
+
"learning_rate": 7.157141191620548e-07,
|
| 393 |
+
"loss": 0.2618,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 11.2,
|
| 398 |
+
"grad_norm": 0.3542459747594626,
|
| 399 |
+
"learning_rate": 5.824397777698859e-07,
|
| 400 |
+
"loss": 0.2768,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 11.4,
|
| 405 |
+
"grad_norm": 0.3647484821206581,
|
| 406 |
+
"learning_rate": 4.6212290164521554e-07,
|
| 407 |
+
"loss": 0.2579,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 11.6,
|
| 412 |
+
"grad_norm": 0.3572146335428771,
|
| 413 |
+
"learning_rate": 3.5511640091604293e-07,
|
| 414 |
+
"loss": 0.2604,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 11.8,
|
| 419 |
+
"grad_norm": 0.3263231215974691,
|
| 420 |
+
"learning_rate": 2.617341440859883e-07,
|
| 421 |
+
"loss": 0.2526,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 12.0,
|
| 426 |
+
"grad_norm": 0.3500512368836611,
|
| 427 |
+
"learning_rate": 1.8225003740388546e-07,
|
| 428 |
+
"loss": 0.2719,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 12.2,
|
| 433 |
+
"grad_norm": 0.3434723627022533,
|
| 434 |
+
"learning_rate": 1.1689722144956672e-07,
|
| 435 |
+
"loss": 0.2767,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 12.4,
|
| 440 |
+
"grad_norm": 0.3535578464793507,
|
| 441 |
+
"learning_rate": 6.58673872923693e-08,
|
| 442 |
+
"loss": 0.2726,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 12.6,
|
| 447 |
+
"grad_norm": 0.3308948888330182,
|
| 448 |
+
"learning_rate": 2.9310214228202016e-08,
|
| 449 |
+
"loss": 0.2583,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 12.8,
|
| 454 |
+
"grad_norm": 0.3515345904747637,
|
| 455 |
+
"learning_rate": 7.332930744380906e-09,
|
| 456 |
+
"loss": 0.262,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 13.0,
|
| 461 |
+
"grad_norm": 0.33983888102120363,
|
| 462 |
+
"learning_rate": 0.0,
|
| 463 |
+
"loss": 0.2516,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 13.0,
|
| 468 |
+
"step": 65,
|
| 469 |
+
"total_flos": 8.525099631326003e+16,
|
| 470 |
+
"train_loss": 0.472858812717291,
|
| 471 |
+
"train_runtime": 5413.2141,
|
| 472 |
+
"train_samples_per_second": 0.759,
|
| 473 |
+
"train_steps_per_second": 0.012
|
| 474 |
+
}
|
| 475 |
+
],
|
| 476 |
+
"logging_steps": 1.0,
|
| 477 |
+
"max_steps": 65,
|
| 478 |
+
"num_input_tokens_seen": 0,
|
| 479 |
+
"num_train_epochs": 13,
|
| 480 |
+
"save_steps": 500,
|
| 481 |
+
"stateful_callbacks": {
|
| 482 |
+
"TrainerControl": {
|
| 483 |
+
"args": {
|
| 484 |
+
"should_epoch_stop": false,
|
| 485 |
+
"should_evaluate": false,
|
| 486 |
+
"should_log": false,
|
| 487 |
+
"should_save": true,
|
| 488 |
+
"should_training_stop": true
|
| 489 |
+
},
|
| 490 |
+
"attributes": {}
|
| 491 |
+
}
|
| 492 |
+
},
|
| 493 |
+
"total_flos": 8.525099631326003e+16,
|
| 494 |
+
"train_batch_size": 1,
|
| 495 |
+
"trial_name": null,
|
| 496 |
+
"trial_params": null
|
| 497 |
+
}
|
training_loss.png
ADDED
|