End of training
Browse files- README.md +2 -1
- all_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +567 -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_lr2e5_epochs15
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# 316_globalbatchsize64_lr2e5_epochs15
|
| 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_lr2e5_epochs15
|
|
|
|
| 16 |
|
| 17 |
# 316_globalbatchsize64_lr2e5_epochs15
|
| 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": 15.0,
|
| 3 |
+
"total_flos": 9.86030068989952e+16,
|
| 4 |
+
"train_loss": 0.28399625040590765,
|
| 5 |
+
"train_runtime": 6242.4075,
|
| 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": 15.0,
|
| 3 |
+
"total_flos": 9.86030068989952e+16,
|
| 4 |
+
"train_loss": 0.28399625040590765,
|
| 5 |
+
"train_runtime": 6242.4075,
|
| 6 |
+
"train_samples_per_second": 0.759,
|
| 7 |
+
"train_steps_per_second": 0.012
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,567 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 15.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 75,
|
| 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": 5.5820740662503425,
|
| 14 |
+
"learning_rate": 2.5e-06,
|
| 15 |
+
"loss": 0.832,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.4,
|
| 20 |
+
"grad_norm": 6.008099573926324,
|
| 21 |
+
"learning_rate": 5e-06,
|
| 22 |
+
"loss": 0.8909,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.6,
|
| 27 |
+
"grad_norm": 5.706742421629664,
|
| 28 |
+
"learning_rate": 7.500000000000001e-06,
|
| 29 |
+
"loss": 0.8975,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.8,
|
| 34 |
+
"grad_norm": 4.258416334954996,
|
| 35 |
+
"learning_rate": 1e-05,
|
| 36 |
+
"loss": 0.8647,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 1.0,
|
| 41 |
+
"grad_norm": 2.0857868158647546,
|
| 42 |
+
"learning_rate": 1.25e-05,
|
| 43 |
+
"loss": 0.7838,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 1.2,
|
| 48 |
+
"grad_norm": 5.647280764828131,
|
| 49 |
+
"learning_rate": 1.5000000000000002e-05,
|
| 50 |
+
"loss": 0.8144,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 1.4,
|
| 55 |
+
"grad_norm": 8.569379397089753,
|
| 56 |
+
"learning_rate": 1.7500000000000002e-05,
|
| 57 |
+
"loss": 0.8566,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 1.6,
|
| 62 |
+
"grad_norm": 7.735280978633036,
|
| 63 |
+
"learning_rate": 2e-05,
|
| 64 |
+
"loss": 0.837,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 1.8,
|
| 69 |
+
"grad_norm": 4.574011055853512,
|
| 70 |
+
"learning_rate": 1.9989008914857115e-05,
|
| 71 |
+
"loss": 0.7907,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 2.0,
|
| 76 |
+
"grad_norm": 3.2366031329888023,
|
| 77 |
+
"learning_rate": 1.9956059820218982e-05,
|
| 78 |
+
"loss": 0.7275,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 2.2,
|
| 83 |
+
"grad_norm": 2.351602239413264,
|
| 84 |
+
"learning_rate": 1.990122514534651e-05,
|
| 85 |
+
"loss": 0.7024,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 2.4,
|
| 90 |
+
"grad_norm": 1.824388674691251,
|
| 91 |
+
"learning_rate": 1.982462542875576e-05,
|
| 92 |
+
"loss": 0.6445,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 2.6,
|
| 97 |
+
"grad_norm": 1.2250296782921972,
|
| 98 |
+
"learning_rate": 1.972642905324813e-05,
|
| 99 |
+
"loss": 0.6175,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 2.8,
|
| 104 |
+
"grad_norm": 1.0940321158989026,
|
| 105 |
+
"learning_rate": 1.9606851875768404e-05,
|
| 106 |
+
"loss": 0.6367,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 3.0,
|
| 111 |
+
"grad_norm": 1.062560084444676,
|
| 112 |
+
"learning_rate": 1.9466156752904344e-05,
|
| 113 |
+
"loss": 0.607,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 3.2,
|
| 118 |
+
"grad_norm": 0.8418790140645667,
|
| 119 |
+
"learning_rate": 1.9304652963070868e-05,
|
| 120 |
+
"loss": 0.523,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 3.4,
|
| 125 |
+
"grad_norm": 0.8596411326553409,
|
| 126 |
+
"learning_rate": 1.9122695526648968e-05,
|
| 127 |
+
"loss": 0.5878,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 3.6,
|
| 132 |
+
"grad_norm": 2.683590177932973,
|
| 133 |
+
"learning_rate": 1.8920684425573865e-05,
|
| 134 |
+
"loss": 0.5315,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 3.8,
|
| 139 |
+
"grad_norm": 1.0546413161380288,
|
| 140 |
+
"learning_rate": 1.8699063724087905e-05,
|
| 141 |
+
"loss": 0.5265,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 4.0,
|
| 146 |
+
"grad_norm": 0.7205640524524933,
|
| 147 |
+
"learning_rate": 1.8458320592590976e-05,
|
| 148 |
+
"loss": 0.4572,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 4.2,
|
| 153 |
+
"grad_norm": 0.8440191277505408,
|
| 154 |
+
"learning_rate": 1.8198984236734246e-05,
|
| 155 |
+
"loss": 0.4667,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 4.4,
|
| 160 |
+
"grad_norm": 0.7832786494693724,
|
| 161 |
+
"learning_rate": 1.7921624734111292e-05,
|
| 162 |
+
"loss": 0.4186,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 4.6,
|
| 167 |
+
"grad_norm": 0.7726547610804203,
|
| 168 |
+
"learning_rate": 1.762685178110382e-05,
|
| 169 |
+
"loss": 0.4149,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 4.8,
|
| 174 |
+
"grad_norm": 0.7438101223901954,
|
| 175 |
+
"learning_rate": 1.731531335263669e-05,
|
| 176 |
+
"loss": 0.4219,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 5.0,
|
| 181 |
+
"grad_norm": 0.9344642511130782,
|
| 182 |
+
"learning_rate": 1.698769427778842e-05,
|
| 183 |
+
"loss": 0.399,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 5.2,
|
| 188 |
+
"grad_norm": 0.8455099974475572,
|
| 189 |
+
"learning_rate": 1.664471473438822e-05,
|
| 190 |
+
"loss": 0.3884,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 5.4,
|
| 195 |
+
"grad_norm": 0.7685465960306714,
|
| 196 |
+
"learning_rate": 1.628712866590885e-05,
|
| 197 |
+
"loss": 0.3403,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 5.6,
|
| 202 |
+
"grad_norm": 0.7248937279387117,
|
| 203 |
+
"learning_rate": 1.5915722124135227e-05,
|
| 204 |
+
"loss": 0.3294,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 5.8,
|
| 209 |
+
"grad_norm": 0.8361135069555778,
|
| 210 |
+
"learning_rate": 1.5531311541251995e-05,
|
| 211 |
+
"loss": 0.2918,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 6.0,
|
| 216 |
+
"grad_norm": 0.6214130974166043,
|
| 217 |
+
"learning_rate": 1.513474193514842e-05,
|
| 218 |
+
"loss": 0.2839,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 6.2,
|
| 223 |
+
"grad_norm": 0.8149573400371725,
|
| 224 |
+
"learning_rate": 1.4726885051885654e-05,
|
| 225 |
+
"loss": 0.2609,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 6.4,
|
| 230 |
+
"grad_norm": 1.0360067290148387,
|
| 231 |
+
"learning_rate": 1.4308637449409705e-05,
|
| 232 |
+
"loss": 0.2481,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 6.6,
|
| 237 |
+
"grad_norm": 0.7209378246648206,
|
| 238 |
+
"learning_rate": 1.3880918526722497e-05,
|
| 239 |
+
"loss": 0.2202,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 6.8,
|
| 244 |
+
"grad_norm": 0.7516647486638462,
|
| 245 |
+
"learning_rate": 1.344466850284333e-05,
|
| 246 |
+
"loss": 0.2246,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 7.0,
|
| 251 |
+
"grad_norm": 0.5980564556653549,
|
| 252 |
+
"learning_rate": 1.300084635000341e-05,
|
| 253 |
+
"loss": 0.2106,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 7.2,
|
| 258 |
+
"grad_norm": 0.7302897741304413,
|
| 259 |
+
"learning_rate": 1.2550427685616767e-05,
|
| 260 |
+
"loss": 0.1665,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 7.4,
|
| 265 |
+
"grad_norm": 0.7955427448866793,
|
| 266 |
+
"learning_rate": 1.2094402627661447e-05,
|
| 267 |
+
"loss": 0.1765,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 7.6,
|
| 272 |
+
"grad_norm": 0.6947342716788957,
|
| 273 |
+
"learning_rate": 1.1633773618185302e-05,
|
| 274 |
+
"loss": 0.1476,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 7.8,
|
| 279 |
+
"grad_norm": 0.666613403120678,
|
| 280 |
+
"learning_rate": 1.1169553219720828e-05,
|
| 281 |
+
"loss": 0.1556,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 8.0,
|
| 286 |
+
"grad_norm": 0.7246014420654745,
|
| 287 |
+
"learning_rate": 1.070276188945293e-05,
|
| 288 |
+
"loss": 0.1497,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 8.2,
|
| 293 |
+
"grad_norm": 0.5352518212418655,
|
| 294 |
+
"learning_rate": 1.0234425736032607e-05,
|
| 295 |
+
"loss": 0.1054,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 8.4,
|
| 300 |
+
"grad_norm": 0.582620097153326,
|
| 301 |
+
"learning_rate": 9.765574263967397e-06,
|
| 302 |
+
"loss": 0.1116,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 8.6,
|
| 307 |
+
"grad_norm": 0.7722198112951959,
|
| 308 |
+
"learning_rate": 9.297238110547075e-06,
|
| 309 |
+
"loss": 0.1009,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 8.8,
|
| 314 |
+
"grad_norm": 0.7769161831075575,
|
| 315 |
+
"learning_rate": 8.830446780279175e-06,
|
| 316 |
+
"loss": 0.1081,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 9.0,
|
| 321 |
+
"grad_norm": 0.5267516336891753,
|
| 322 |
+
"learning_rate": 8.366226381814698e-06,
|
| 323 |
+
"loss": 0.0924,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 9.2,
|
| 328 |
+
"grad_norm": 0.6815118378301678,
|
| 329 |
+
"learning_rate": 7.905597372338558e-06,
|
| 330 |
+
"loss": 0.0788,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 9.4,
|
| 335 |
+
"grad_norm": 0.6096654056601541,
|
| 336 |
+
"learning_rate": 7.449572314383237e-06,
|
| 337 |
+
"loss": 0.0704,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 9.6,
|
| 342 |
+
"grad_norm": 0.4975893608049929,
|
| 343 |
+
"learning_rate": 6.999153649996595e-06,
|
| 344 |
+
"loss": 0.0608,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 9.8,
|
| 349 |
+
"grad_norm": 0.5089023783497421,
|
| 350 |
+
"learning_rate": 6.555331497156671e-06,
|
| 351 |
+
"loss": 0.0637,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 10.0,
|
| 356 |
+
"grad_norm": 0.7005654148661791,
|
| 357 |
+
"learning_rate": 6.119081473277502e-06,
|
| 358 |
+
"loss": 0.0613,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 10.2,
|
| 363 |
+
"grad_norm": 0.46142897491616885,
|
| 364 |
+
"learning_rate": 5.6913625505902966e-06,
|
| 365 |
+
"loss": 0.0411,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 10.4,
|
| 370 |
+
"grad_norm": 0.3561622014470182,
|
| 371 |
+
"learning_rate": 5.273114948114346e-06,
|
| 372 |
+
"loss": 0.0455,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 10.6,
|
| 377 |
+
"grad_norm": 0.35910813816061127,
|
| 378 |
+
"learning_rate": 4.865258064851579e-06,
|
| 379 |
+
"loss": 0.0448,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 10.8,
|
| 384 |
+
"grad_norm": 0.3319709763565861,
|
| 385 |
+
"learning_rate": 4.468688458748006e-06,
|
| 386 |
+
"loss": 0.0374,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 11.0,
|
| 391 |
+
"grad_norm": 0.33083801499691196,
|
| 392 |
+
"learning_rate": 4.084277875864776e-06,
|
| 393 |
+
"loss": 0.0364,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 11.2,
|
| 398 |
+
"grad_norm": 0.321908630777182,
|
| 399 |
+
"learning_rate": 3.712871334091154e-06,
|
| 400 |
+
"loss": 0.0324,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 11.4,
|
| 405 |
+
"grad_norm": 0.28605671407511096,
|
| 406 |
+
"learning_rate": 3.355285265611784e-06,
|
| 407 |
+
"loss": 0.0277,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 11.6,
|
| 412 |
+
"grad_norm": 0.2613134954891153,
|
| 413 |
+
"learning_rate": 3.0123057222115835e-06,
|
| 414 |
+
"loss": 0.0254,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 11.8,
|
| 419 |
+
"grad_norm": 0.26765400526301425,
|
| 420 |
+
"learning_rate": 2.6846866473633126e-06,
|
| 421 |
+
"loss": 0.0252,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 12.0,
|
| 426 |
+
"grad_norm": 0.24309760367630573,
|
| 427 |
+
"learning_rate": 2.373148218896182e-06,
|
| 428 |
+
"loss": 0.0264,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 12.2,
|
| 433 |
+
"grad_norm": 0.20368260468560698,
|
| 434 |
+
"learning_rate": 2.078375265888707e-06,
|
| 435 |
+
"loss": 0.0245,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 12.4,
|
| 440 |
+
"grad_norm": 0.23111592875956644,
|
| 441 |
+
"learning_rate": 1.8010157632657544e-06,
|
| 442 |
+
"loss": 0.0217,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 12.6,
|
| 447 |
+
"grad_norm": 0.21424368315616057,
|
| 448 |
+
"learning_rate": 1.5416794074090258e-06,
|
| 449 |
+
"loss": 0.02,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 12.8,
|
| 454 |
+
"grad_norm": 0.23486365442991192,
|
| 455 |
+
"learning_rate": 1.300936275912098e-06,
|
| 456 |
+
"loss": 0.0204,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 13.0,
|
| 461 |
+
"grad_norm": 0.1966413751838419,
|
| 462 |
+
"learning_rate": 1.0793155744261352e-06,
|
| 463 |
+
"loss": 0.0162,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 13.2,
|
| 468 |
+
"grad_norm": 0.1590508395588302,
|
| 469 |
+
"learning_rate": 8.773044733510338e-07,
|
| 470 |
+
"loss": 0.0181,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 13.4,
|
| 475 |
+
"grad_norm": 0.14626626011540572,
|
| 476 |
+
"learning_rate": 6.953470369291349e-07,
|
| 477 |
+
"loss": 0.0143,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 13.6,
|
| 482 |
+
"grad_norm": 0.16591443824765065,
|
| 483 |
+
"learning_rate": 5.33843247095659e-07,
|
| 484 |
+
"loss": 0.0194,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 13.8,
|
| 489 |
+
"grad_norm": 0.1709470448469729,
|
| 490 |
+
"learning_rate": 3.931481242315993e-07,
|
| 491 |
+
"loss": 0.0168,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 14.0,
|
| 496 |
+
"grad_norm": 0.14577825961665566,
|
| 497 |
+
"learning_rate": 2.735709467518699e-07,
|
| 498 |
+
"loss": 0.0134,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 14.2,
|
| 503 |
+
"grad_norm": 0.13945121547728515,
|
| 504 |
+
"learning_rate": 1.7537457124423896e-07,
|
| 505 |
+
"loss": 0.0134,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 14.4,
|
| 510 |
+
"grad_norm": 0.14310398445495884,
|
| 511 |
+
"learning_rate": 9.877485465349057e-08,
|
| 512 |
+
"loss": 0.0152,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 14.6,
|
| 517 |
+
"grad_norm": 0.1404230436657977,
|
| 518 |
+
"learning_rate": 4.394017978101905e-08,
|
| 519 |
+
"loss": 0.0137,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 14.8,
|
| 524 |
+
"grad_norm": 0.16581805815771083,
|
| 525 |
+
"learning_rate": 1.099108514288627e-08,
|
| 526 |
+
"loss": 0.0161,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 15.0,
|
| 531 |
+
"grad_norm": 0.14366857308957606,
|
| 532 |
+
"learning_rate": 0.0,
|
| 533 |
+
"loss": 0.0168,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 15.0,
|
| 538 |
+
"step": 75,
|
| 539 |
+
"total_flos": 9.86030068989952e+16,
|
| 540 |
+
"train_loss": 0.28399625040590765,
|
| 541 |
+
"train_runtime": 6242.4075,
|
| 542 |
+
"train_samples_per_second": 0.759,
|
| 543 |
+
"train_steps_per_second": 0.012
|
| 544 |
+
}
|
| 545 |
+
],
|
| 546 |
+
"logging_steps": 1.0,
|
| 547 |
+
"max_steps": 75,
|
| 548 |
+
"num_input_tokens_seen": 0,
|
| 549 |
+
"num_train_epochs": 15,
|
| 550 |
+
"save_steps": 500,
|
| 551 |
+
"stateful_callbacks": {
|
| 552 |
+
"TrainerControl": {
|
| 553 |
+
"args": {
|
| 554 |
+
"should_epoch_stop": false,
|
| 555 |
+
"should_evaluate": false,
|
| 556 |
+
"should_log": false,
|
| 557 |
+
"should_save": true,
|
| 558 |
+
"should_training_stop": true
|
| 559 |
+
},
|
| 560 |
+
"attributes": {}
|
| 561 |
+
}
|
| 562 |
+
},
|
| 563 |
+
"total_flos": 9.86030068989952e+16,
|
| 564 |
+
"train_batch_size": 1,
|
| 565 |
+
"trial_name": null,
|
| 566 |
+
"trial_params": null
|
| 567 |
+
}
|
training_loss.png
ADDED
|