Commit
•
021eaef
1
Parent(s):
8b44b83
End of training
Browse files- pytorch_model.bin +1 -1
- tokenizer_config.json +2 -0
- trainer_state.json +688 -6
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 139279005
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3b7c8819f663fe4a7adc490126b907fbe6357c071df3563631ef43908f2bb223
|
| 3 |
size 139279005
|
tokenizer_config.json
CHANGED
|
@@ -1,4 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"model_max_length": 1000000000000000019884624838656,
|
|
|
|
|
|
|
| 3 |
"tokenizer_class": "PreTrainedTokenizerFast"
|
| 4 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"model_max_length": 1000000000000000019884624838656,
|
| 3 |
+
"name_or_path": "JammyMachina/improved_4bars-mdl",
|
| 4 |
+
"special_tokens_map_file": "/root/.cache/huggingface/hub/models--JammyMachina--improved_4bars-mdl/snapshots/6498a0943c2a7f99cc4944c22daf716e6de22f38/special_tokens_map.json",
|
| 5 |
"tokenizer_class": "PreTrainedTokenizerFast"
|
| 6 |
}
|
trainer_state.json
CHANGED
|
@@ -1,15 +1,697 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
-
"global_step":
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
-
"log_history": [
|
| 10 |
-
|
| 11 |
-
|
| 12 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 13 |
"trial_name": null,
|
| 14 |
"trial_params": null
|
| 15 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
+
"global_step": 49420,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.1,
|
| 12 |
+
"learning_rate": 4.896398219344395e-05,
|
| 13 |
+
"loss": 0.2574,
|
| 14 |
+
"step": 1024
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.1,
|
| 18 |
+
"eval_loss": 0.7888669967651367,
|
| 19 |
+
"eval_runtime": 25.6574,
|
| 20 |
+
"eval_samples_per_second": 81.536,
|
| 21 |
+
"eval_steps_per_second": 3.43,
|
| 22 |
+
"step": 1024
|
| 23 |
+
},
|
| 24 |
+
{
|
| 25 |
+
"epoch": 0.21,
|
| 26 |
+
"learning_rate": 4.79279643868879e-05,
|
| 27 |
+
"loss": 0.2633,
|
| 28 |
+
"step": 2048
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 0.21,
|
| 32 |
+
"eval_loss": 0.780219316482544,
|
| 33 |
+
"eval_runtime": 25.6023,
|
| 34 |
+
"eval_samples_per_second": 81.711,
|
| 35 |
+
"eval_steps_per_second": 3.437,
|
| 36 |
+
"step": 2048
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 0.31,
|
| 40 |
+
"learning_rate": 4.6891946580331854e-05,
|
| 41 |
+
"loss": 0.2635,
|
| 42 |
+
"step": 3072
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 0.31,
|
| 46 |
+
"eval_loss": 0.7877150774002075,
|
| 47 |
+
"eval_runtime": 25.6162,
|
| 48 |
+
"eval_samples_per_second": 81.667,
|
| 49 |
+
"eval_steps_per_second": 3.435,
|
| 50 |
+
"step": 3072
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.41,
|
| 54 |
+
"learning_rate": 4.5855928773775804e-05,
|
| 55 |
+
"loss": 0.2639,
|
| 56 |
+
"step": 4096
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.41,
|
| 60 |
+
"eval_loss": 0.7751328945159912,
|
| 61 |
+
"eval_runtime": 25.6157,
|
| 62 |
+
"eval_samples_per_second": 81.669,
|
| 63 |
+
"eval_steps_per_second": 3.435,
|
| 64 |
+
"step": 4096
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.52,
|
| 68 |
+
"learning_rate": 4.4819910967219755e-05,
|
| 69 |
+
"loss": 0.2625,
|
| 70 |
+
"step": 5120
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 0.52,
|
| 74 |
+
"eval_loss": 0.7836451530456543,
|
| 75 |
+
"eval_runtime": 25.6812,
|
| 76 |
+
"eval_samples_per_second": 81.46,
|
| 77 |
+
"eval_steps_per_second": 3.427,
|
| 78 |
+
"step": 5120
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 0.62,
|
| 82 |
+
"learning_rate": 4.3783893160663705e-05,
|
| 83 |
+
"loss": 0.2609,
|
| 84 |
+
"step": 6144
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.62,
|
| 88 |
+
"eval_loss": 0.7758111357688904,
|
| 89 |
+
"eval_runtime": 25.6349,
|
| 90 |
+
"eval_samples_per_second": 81.607,
|
| 91 |
+
"eval_steps_per_second": 3.433,
|
| 92 |
+
"step": 6144
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.73,
|
| 96 |
+
"learning_rate": 4.2747875354107656e-05,
|
| 97 |
+
"loss": 0.2597,
|
| 98 |
+
"step": 7168
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.73,
|
| 102 |
+
"eval_loss": 0.7923067212104797,
|
| 103 |
+
"eval_runtime": 25.6218,
|
| 104 |
+
"eval_samples_per_second": 81.649,
|
| 105 |
+
"eval_steps_per_second": 3.435,
|
| 106 |
+
"step": 7168
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.83,
|
| 110 |
+
"learning_rate": 4.17118575475516e-05,
|
| 111 |
+
"loss": 0.2612,
|
| 112 |
+
"step": 8192
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 0.83,
|
| 116 |
+
"eval_loss": 0.8000349402427673,
|
| 117 |
+
"eval_runtime": 25.646,
|
| 118 |
+
"eval_samples_per_second": 81.572,
|
| 119 |
+
"eval_steps_per_second": 3.431,
|
| 120 |
+
"step": 8192
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 0.93,
|
| 124 |
+
"learning_rate": 4.0676851477134767e-05,
|
| 125 |
+
"loss": 0.2612,
|
| 126 |
+
"step": 9216
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 0.93,
|
| 130 |
+
"eval_loss": 0.7934667468070984,
|
| 131 |
+
"eval_runtime": 25.6816,
|
| 132 |
+
"eval_samples_per_second": 81.459,
|
| 133 |
+
"eval_steps_per_second": 3.427,
|
| 134 |
+
"step": 9216
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 1.04,
|
| 138 |
+
"learning_rate": 3.964083367057872e-05,
|
| 139 |
+
"loss": 0.2554,
|
| 140 |
+
"step": 10240
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 1.04,
|
| 144 |
+
"eval_loss": 0.7942602038383484,
|
| 145 |
+
"eval_runtime": 25.7479,
|
| 146 |
+
"eval_samples_per_second": 81.249,
|
| 147 |
+
"eval_steps_per_second": 3.418,
|
| 148 |
+
"step": 10240
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 1.14,
|
| 152 |
+
"learning_rate": 3.860481586402267e-05,
|
| 153 |
+
"loss": 0.2524,
|
| 154 |
+
"step": 11264
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 1.14,
|
| 158 |
+
"eval_loss": 0.8014633655548096,
|
| 159 |
+
"eval_runtime": 25.6355,
|
| 160 |
+
"eval_samples_per_second": 81.606,
|
| 161 |
+
"eval_steps_per_second": 3.433,
|
| 162 |
+
"step": 11264
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 1.24,
|
| 166 |
+
"learning_rate": 3.7569809793605834e-05,
|
| 167 |
+
"loss": 0.2504,
|
| 168 |
+
"step": 12288
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 1.24,
|
| 172 |
+
"eval_loss": 0.7962351441383362,
|
| 173 |
+
"eval_runtime": 25.6019,
|
| 174 |
+
"eval_samples_per_second": 81.713,
|
| 175 |
+
"eval_steps_per_second": 3.437,
|
| 176 |
+
"step": 12288
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 1.35,
|
| 180 |
+
"learning_rate": 3.653379198704978e-05,
|
| 181 |
+
"loss": 0.2524,
|
| 182 |
+
"step": 13312
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 1.35,
|
| 186 |
+
"eval_loss": 0.8086119294166565,
|
| 187 |
+
"eval_runtime": 25.7264,
|
| 188 |
+
"eval_samples_per_second": 81.317,
|
| 189 |
+
"eval_steps_per_second": 3.421,
|
| 190 |
+
"step": 13312
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 1.45,
|
| 194 |
+
"learning_rate": 3.549777418049373e-05,
|
| 195 |
+
"loss": 0.2521,
|
| 196 |
+
"step": 14336
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 1.45,
|
| 200 |
+
"eval_loss": 0.8062009215354919,
|
| 201 |
+
"eval_runtime": 25.6245,
|
| 202 |
+
"eval_samples_per_second": 81.641,
|
| 203 |
+
"eval_steps_per_second": 3.434,
|
| 204 |
+
"step": 14336
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 1.55,
|
| 208 |
+
"learning_rate": 3.446175637393768e-05,
|
| 209 |
+
"loss": 0.2503,
|
| 210 |
+
"step": 15360
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 1.55,
|
| 214 |
+
"eval_loss": 0.7997719049453735,
|
| 215 |
+
"eval_runtime": 25.6588,
|
| 216 |
+
"eval_samples_per_second": 81.532,
|
| 217 |
+
"eval_steps_per_second": 3.43,
|
| 218 |
+
"step": 15360
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 1.66,
|
| 222 |
+
"learning_rate": 3.3426750303520846e-05,
|
| 223 |
+
"loss": 0.2523,
|
| 224 |
+
"step": 16384
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 1.66,
|
| 228 |
+
"eval_loss": 0.8097875118255615,
|
| 229 |
+
"eval_runtime": 25.6111,
|
| 230 |
+
"eval_samples_per_second": 81.683,
|
| 231 |
+
"eval_steps_per_second": 3.436,
|
| 232 |
+
"step": 16384
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 1.76,
|
| 236 |
+
"learning_rate": 3.239073249696479e-05,
|
| 237 |
+
"loss": 0.251,
|
| 238 |
+
"step": 17408
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 1.76,
|
| 242 |
+
"eval_loss": 0.8213343620300293,
|
| 243 |
+
"eval_runtime": 25.6868,
|
| 244 |
+
"eval_samples_per_second": 81.443,
|
| 245 |
+
"eval_steps_per_second": 3.426,
|
| 246 |
+
"step": 17408
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 1.86,
|
| 250 |
+
"learning_rate": 3.135572642654796e-05,
|
| 251 |
+
"loss": 0.2509,
|
| 252 |
+
"step": 18432
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 1.86,
|
| 256 |
+
"eval_loss": 0.8137653470039368,
|
| 257 |
+
"eval_runtime": 25.6223,
|
| 258 |
+
"eval_samples_per_second": 81.648,
|
| 259 |
+
"eval_steps_per_second": 3.435,
|
| 260 |
+
"step": 18432
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 1.97,
|
| 264 |
+
"learning_rate": 3.031970861999191e-05,
|
| 265 |
+
"loss": 0.2533,
|
| 266 |
+
"step": 19456
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 1.97,
|
| 270 |
+
"eval_loss": 0.8182128071784973,
|
| 271 |
+
"eval_runtime": 25.6272,
|
| 272 |
+
"eval_samples_per_second": 81.632,
|
| 273 |
+
"eval_steps_per_second": 3.434,
|
| 274 |
+
"step": 19456
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 2.07,
|
| 278 |
+
"learning_rate": 2.928369081343586e-05,
|
| 279 |
+
"loss": 0.245,
|
| 280 |
+
"step": 20480
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 2.07,
|
| 284 |
+
"eval_loss": 0.8290057182312012,
|
| 285 |
+
"eval_runtime": 25.6288,
|
| 286 |
+
"eval_samples_per_second": 81.627,
|
| 287 |
+
"eval_steps_per_second": 3.434,
|
| 288 |
+
"step": 20480
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 2.18,
|
| 292 |
+
"learning_rate": 2.824868474301902e-05,
|
| 293 |
+
"loss": 0.2432,
|
| 294 |
+
"step": 21504
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 2.18,
|
| 298 |
+
"eval_loss": 0.8328016400337219,
|
| 299 |
+
"eval_runtime": 25.7111,
|
| 300 |
+
"eval_samples_per_second": 81.366,
|
| 301 |
+
"eval_steps_per_second": 3.423,
|
| 302 |
+
"step": 21504
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 2.28,
|
| 306 |
+
"learning_rate": 2.7212666936462972e-05,
|
| 307 |
+
"loss": 0.2435,
|
| 308 |
+
"step": 22528
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 2.28,
|
| 312 |
+
"eval_loss": 0.8186773657798767,
|
| 313 |
+
"eval_runtime": 25.61,
|
| 314 |
+
"eval_samples_per_second": 81.687,
|
| 315 |
+
"eval_steps_per_second": 3.436,
|
| 316 |
+
"step": 22528
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 2.38,
|
| 320 |
+
"learning_rate": 2.617766086604614e-05,
|
| 321 |
+
"loss": 0.2423,
|
| 322 |
+
"step": 23552
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 2.38,
|
| 326 |
+
"eval_loss": 0.8238145112991333,
|
| 327 |
+
"eval_runtime": 25.6287,
|
| 328 |
+
"eval_samples_per_second": 81.627,
|
| 329 |
+
"eval_steps_per_second": 3.434,
|
| 330 |
+
"step": 23552
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 2.49,
|
| 334 |
+
"learning_rate": 2.51426547956293e-05,
|
| 335 |
+
"loss": 0.2443,
|
| 336 |
+
"step": 24576
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 2.49,
|
| 340 |
+
"eval_loss": 0.8248941898345947,
|
| 341 |
+
"eval_runtime": 25.6771,
|
| 342 |
+
"eval_samples_per_second": 81.473,
|
| 343 |
+
"eval_steps_per_second": 3.427,
|
| 344 |
+
"step": 24576
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 2.59,
|
| 348 |
+
"learning_rate": 2.410663698907325e-05,
|
| 349 |
+
"loss": 0.2431,
|
| 350 |
+
"step": 25600
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 2.59,
|
| 354 |
+
"eval_loss": 0.8253324031829834,
|
| 355 |
+
"eval_runtime": 25.6574,
|
| 356 |
+
"eval_samples_per_second": 81.536,
|
| 357 |
+
"eval_steps_per_second": 3.43,
|
| 358 |
+
"step": 25600
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 2.69,
|
| 362 |
+
"learning_rate": 2.30706191825172e-05,
|
| 363 |
+
"loss": 0.2432,
|
| 364 |
+
"step": 26624
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 2.69,
|
| 368 |
+
"eval_loss": 0.8268725275993347,
|
| 369 |
+
"eval_runtime": 25.6215,
|
| 370 |
+
"eval_samples_per_second": 81.65,
|
| 371 |
+
"eval_steps_per_second": 3.435,
|
| 372 |
+
"step": 26624
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 2.8,
|
| 376 |
+
"learning_rate": 2.203460137596115e-05,
|
| 377 |
+
"loss": 0.2421,
|
| 378 |
+
"step": 27648
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 2.8,
|
| 382 |
+
"eval_loss": 0.8282283544540405,
|
| 383 |
+
"eval_runtime": 25.6027,
|
| 384 |
+
"eval_samples_per_second": 81.71,
|
| 385 |
+
"eval_steps_per_second": 3.437,
|
| 386 |
+
"step": 27648
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 2.9,
|
| 390 |
+
"learning_rate": 2.09985835694051e-05,
|
| 391 |
+
"loss": 0.2421,
|
| 392 |
+
"step": 28672
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 2.9,
|
| 396 |
+
"eval_loss": 0.8267728686332703,
|
| 397 |
+
"eval_runtime": 25.6314,
|
| 398 |
+
"eval_samples_per_second": 81.619,
|
| 399 |
+
"eval_steps_per_second": 3.433,
|
| 400 |
+
"step": 28672
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 3.0,
|
| 404 |
+
"learning_rate": 1.9963577498988265e-05,
|
| 405 |
+
"loss": 0.243,
|
| 406 |
+
"step": 29696
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 3.0,
|
| 410 |
+
"eval_loss": 0.8345196843147278,
|
| 411 |
+
"eval_runtime": 25.6765,
|
| 412 |
+
"eval_samples_per_second": 81.475,
|
| 413 |
+
"eval_steps_per_second": 3.427,
|
| 414 |
+
"step": 29696
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 3.11,
|
| 418 |
+
"learning_rate": 1.8927559692432215e-05,
|
| 419 |
+
"loss": 0.2367,
|
| 420 |
+
"step": 30720
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 3.11,
|
| 424 |
+
"eval_loss": 0.8424174189567566,
|
| 425 |
+
"eval_runtime": 25.634,
|
| 426 |
+
"eval_samples_per_second": 81.61,
|
| 427 |
+
"eval_steps_per_second": 3.433,
|
| 428 |
+
"step": 30720
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 3.21,
|
| 432 |
+
"learning_rate": 1.789255362201538e-05,
|
| 433 |
+
"loss": 0.237,
|
| 434 |
+
"step": 31744
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 3.21,
|
| 438 |
+
"eval_loss": 0.8373659253120422,
|
| 439 |
+
"eval_runtime": 25.5978,
|
| 440 |
+
"eval_samples_per_second": 81.726,
|
| 441 |
+
"eval_steps_per_second": 3.438,
|
| 442 |
+
"step": 31744
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 3.32,
|
| 446 |
+
"learning_rate": 1.6857547551598542e-05,
|
| 447 |
+
"loss": 0.2351,
|
| 448 |
+
"step": 32768
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 3.32,
|
| 452 |
+
"eval_loss": 0.8430664539337158,
|
| 453 |
+
"eval_runtime": 25.6675,
|
| 454 |
+
"eval_samples_per_second": 81.504,
|
| 455 |
+
"eval_steps_per_second": 3.428,
|
| 456 |
+
"step": 32768
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 3.42,
|
| 460 |
+
"learning_rate": 1.5821529745042493e-05,
|
| 461 |
+
"loss": 0.2374,
|
| 462 |
+
"step": 33792
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 3.42,
|
| 466 |
+
"eval_loss": 0.8425076603889465,
|
| 467 |
+
"eval_runtime": 25.7071,
|
| 468 |
+
"eval_samples_per_second": 81.378,
|
| 469 |
+
"eval_steps_per_second": 3.423,
|
| 470 |
+
"step": 33792
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 3.52,
|
| 474 |
+
"learning_rate": 1.4786523674625658e-05,
|
| 475 |
+
"loss": 0.2355,
|
| 476 |
+
"step": 34816
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 3.52,
|
| 480 |
+
"eval_loss": 0.8351919054985046,
|
| 481 |
+
"eval_runtime": 25.6245,
|
| 482 |
+
"eval_samples_per_second": 81.641,
|
| 483 |
+
"eval_steps_per_second": 3.434,
|
| 484 |
+
"step": 34816
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 3.63,
|
| 488 |
+
"learning_rate": 1.3750505868069607e-05,
|
| 489 |
+
"loss": 0.2373,
|
| 490 |
+
"step": 35840
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 3.63,
|
| 494 |
+
"eval_loss": 0.845241904258728,
|
| 495 |
+
"eval_runtime": 25.6392,
|
| 496 |
+
"eval_samples_per_second": 81.594,
|
| 497 |
+
"eval_steps_per_second": 3.432,
|
| 498 |
+
"step": 35840
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 3.73,
|
| 502 |
+
"learning_rate": 1.2714488061513558e-05,
|
| 503 |
+
"loss": 0.2356,
|
| 504 |
+
"step": 36864
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 3.73,
|
| 508 |
+
"eval_loss": 0.8382672071456909,
|
| 509 |
+
"eval_runtime": 25.6391,
|
| 510 |
+
"eval_samples_per_second": 81.594,
|
| 511 |
+
"eval_steps_per_second": 3.432,
|
| 512 |
+
"step": 36864
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 3.83,
|
| 516 |
+
"learning_rate": 1.1678470254957508e-05,
|
| 517 |
+
"loss": 0.2343,
|
| 518 |
+
"step": 37888
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 3.83,
|
| 522 |
+
"eval_loss": 0.8443974852561951,
|
| 523 |
+
"eval_runtime": 25.7018,
|
| 524 |
+
"eval_samples_per_second": 81.395,
|
| 525 |
+
"eval_steps_per_second": 3.424,
|
| 526 |
+
"step": 37888
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 3.94,
|
| 530 |
+
"learning_rate": 1.0642452448401457e-05,
|
| 531 |
+
"loss": 0.2348,
|
| 532 |
+
"step": 38912
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 3.94,
|
| 536 |
+
"eval_loss": 0.8427897095680237,
|
| 537 |
+
"eval_runtime": 25.6146,
|
| 538 |
+
"eval_samples_per_second": 81.672,
|
| 539 |
+
"eval_steps_per_second": 3.436,
|
| 540 |
+
"step": 38912
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 4.04,
|
| 544 |
+
"learning_rate": 9.606434641845407e-06,
|
| 545 |
+
"loss": 0.2349,
|
| 546 |
+
"step": 39936
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 4.04,
|
| 550 |
+
"eval_loss": 0.8479835987091064,
|
| 551 |
+
"eval_runtime": 25.6232,
|
| 552 |
+
"eval_samples_per_second": 81.645,
|
| 553 |
+
"eval_steps_per_second": 3.434,
|
| 554 |
+
"step": 39936
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 4.14,
|
| 558 |
+
"learning_rate": 8.571428571428573e-06,
|
| 559 |
+
"loss": 0.2327,
|
| 560 |
+
"step": 40960
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 4.14,
|
| 564 |
+
"eval_loss": 0.8481870889663696,
|
| 565 |
+
"eval_runtime": 25.6261,
|
| 566 |
+
"eval_samples_per_second": 81.636,
|
| 567 |
+
"eval_steps_per_second": 3.434,
|
| 568 |
+
"step": 40960
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 4.25,
|
| 572 |
+
"learning_rate": 7.536422501011736e-06,
|
| 573 |
+
"loss": 0.2337,
|
| 574 |
+
"step": 41984
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 4.25,
|
| 578 |
+
"eval_loss": 0.850956916809082,
|
| 579 |
+
"eval_runtime": 25.6867,
|
| 580 |
+
"eval_samples_per_second": 81.443,
|
| 581 |
+
"eval_steps_per_second": 3.426,
|
| 582 |
+
"step": 41984
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 4.35,
|
| 586 |
+
"learning_rate": 6.500404694455686e-06,
|
| 587 |
+
"loss": 0.2288,
|
| 588 |
+
"step": 43008
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 4.35,
|
| 592 |
+
"eval_loss": 0.8499058485031128,
|
| 593 |
+
"eval_runtime": 25.6219,
|
| 594 |
+
"eval_samples_per_second": 81.649,
|
| 595 |
+
"eval_steps_per_second": 3.435,
|
| 596 |
+
"step": 43008
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 4.45,
|
| 600 |
+
"learning_rate": 5.464386887899636e-06,
|
| 601 |
+
"loss": 0.2299,
|
| 602 |
+
"step": 44032
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 4.45,
|
| 606 |
+
"eval_loss": 0.8522343039512634,
|
| 607 |
+
"eval_runtime": 25.6037,
|
| 608 |
+
"eval_samples_per_second": 81.707,
|
| 609 |
+
"eval_steps_per_second": 3.437,
|
| 610 |
+
"step": 44032
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 4.56,
|
| 614 |
+
"learning_rate": 4.429380817482801e-06,
|
| 615 |
+
"loss": 0.2277,
|
| 616 |
+
"step": 45056
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 4.56,
|
| 620 |
+
"eval_loss": 0.85256427526474,
|
| 621 |
+
"eval_runtime": 25.6116,
|
| 622 |
+
"eval_samples_per_second": 81.682,
|
| 623 |
+
"eval_steps_per_second": 3.436,
|
| 624 |
+
"step": 45056
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 4.66,
|
| 628 |
+
"learning_rate": 3.3933630109267505e-06,
|
| 629 |
+
"loss": 0.2301,
|
| 630 |
+
"step": 46080
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 4.66,
|
| 634 |
+
"eval_loss": 0.851773202419281,
|
| 635 |
+
"eval_runtime": 25.6644,
|
| 636 |
+
"eval_samples_per_second": 81.514,
|
| 637 |
+
"eval_steps_per_second": 3.429,
|
| 638 |
+
"step": 46080
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 4.77,
|
| 642 |
+
"learning_rate": 2.3573452043707e-06,
|
| 643 |
+
"loss": 0.2312,
|
| 644 |
+
"step": 47104
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 4.77,
|
| 648 |
+
"eval_loss": 0.8511470556259155,
|
| 649 |
+
"eval_runtime": 25.6121,
|
| 650 |
+
"eval_samples_per_second": 81.68,
|
| 651 |
+
"eval_steps_per_second": 3.436,
|
| 652 |
+
"step": 47104
|
| 653 |
+
},
|
| 654 |
+
{
|
| 655 |
+
"epoch": 4.87,
|
| 656 |
+
"learning_rate": 1.32132739781465e-06,
|
| 657 |
+
"loss": 0.2284,
|
| 658 |
+
"step": 48128
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 4.87,
|
| 662 |
+
"eval_loss": 0.8506836295127869,
|
| 663 |
+
"eval_runtime": 25.6413,
|
| 664 |
+
"eval_samples_per_second": 81.587,
|
| 665 |
+
"eval_steps_per_second": 3.432,
|
| 666 |
+
"step": 48128
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 4.97,
|
| 670 |
+
"learning_rate": 2.8632132739781467e-07,
|
| 671 |
+
"loss": 0.2294,
|
| 672 |
+
"step": 49152
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 4.97,
|
| 676 |
+
"eval_loss": 0.8519254922866821,
|
| 677 |
+
"eval_runtime": 25.6136,
|
| 678 |
+
"eval_samples_per_second": 81.675,
|
| 679 |
+
"eval_steps_per_second": 3.436,
|
| 680 |
+
"step": 49152
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 5.0,
|
| 684 |
+
"step": 49420,
|
| 685 |
+
"total_flos": 3.1008084491562394e+17,
|
| 686 |
+
"train_loss": 0.244437825115332,
|
| 687 |
+
"train_runtime": 40418.7139,
|
| 688 |
+
"train_samples_per_second": 25.676,
|
| 689 |
+
"train_steps_per_second": 1.223
|
| 690 |
+
}
|
| 691 |
+
],
|
| 692 |
+
"max_steps": 49420,
|
| 693 |
+
"num_train_epochs": 5,
|
| 694 |
+
"total_flos": 3.1008084491562394e+17,
|
| 695 |
"trial_name": null,
|
| 696 |
"trial_params": null
|
| 697 |
}
|