End of training
Browse files- README.md +15 -2
- all_results.json +15 -0
- eval_results.json +10 -0
- train_results.json +9 -0
- trainer_state.json +417 -0
README.md
CHANGED
|
@@ -3,11 +3,24 @@ library_name: transformers
|
|
| 3 |
base_model: bowphs/pythia-70m-multi
|
| 4 |
tags:
|
| 5 |
- generated_from_trainer
|
|
|
|
|
|
|
| 6 |
metrics:
|
| 7 |
- accuracy
|
| 8 |
model-index:
|
| 9 |
- name: c4-model
|
| 10 |
-
results:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 11 |
---
|
| 12 |
|
| 13 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -15,7 +28,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# c4-model
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [bowphs/pythia-70m-multi](https://huggingface.co/bowphs/pythia-70m-multi) on
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
- Loss: 3.5532
|
| 21 |
- Accuracy: 0.3716
|
|
|
|
| 3 |
base_model: bowphs/pythia-70m-multi
|
| 4 |
tags:
|
| 5 |
- generated_from_trainer
|
| 6 |
+
datasets:
|
| 7 |
+
- allenai/c4
|
| 8 |
metrics:
|
| 9 |
- accuracy
|
| 10 |
model-index:
|
| 11 |
- name: c4-model
|
| 12 |
+
results:
|
| 13 |
+
- task:
|
| 14 |
+
name: Causal Language Modeling
|
| 15 |
+
type: text-generation
|
| 16 |
+
dataset:
|
| 17 |
+
name: allenai/c4 en
|
| 18 |
+
type: allenai/c4
|
| 19 |
+
args: en
|
| 20 |
+
metrics:
|
| 21 |
+
- name: Accuracy
|
| 22 |
+
type: accuracy
|
| 23 |
+
value: 0.3716248289345064
|
| 24 |
---
|
| 25 |
|
| 26 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 28 |
|
| 29 |
# c4-model
|
| 30 |
|
| 31 |
+
This model is a fine-tuned version of [bowphs/pythia-70m-multi](https://huggingface.co/bowphs/pythia-70m-multi) on the allenai/c4 en dataset.
|
| 32 |
It achieves the following results on the evaluation set:
|
| 33 |
- Loss: 3.5532
|
| 34 |
- Accuracy: 0.3716
|
all_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_accuracy": 0.3716248289345064,
|
| 4 |
+
"eval_loss": 3.5532469749450684,
|
| 5 |
+
"eval_runtime": 53.6916,
|
| 6 |
+
"eval_samples_per_second": 93.124,
|
| 7 |
+
"eval_steps_per_second": 2.924,
|
| 8 |
+
"perplexity": 34.92653917175159,
|
| 9 |
+
"total_flos": 2.6348013748224e+17,
|
| 10 |
+
"train_loss": 3.938131396484375,
|
| 11 |
+
"train_runtime": 22585.5788,
|
| 12 |
+
"train_samples": 960000,
|
| 13 |
+
"train_samples_per_second": 42.505,
|
| 14 |
+
"train_steps_per_second": 1.328
|
| 15 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_accuracy": 0.3716248289345064,
|
| 4 |
+
"eval_loss": 3.5532469749450684,
|
| 5 |
+
"eval_runtime": 53.6916,
|
| 6 |
+
"eval_samples_per_second": 93.124,
|
| 7 |
+
"eval_steps_per_second": 2.924,
|
| 8 |
+
"perplexity": 34.92653917175159,
|
| 9 |
+
"train_samples": 960000
|
| 10 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 2.6348013748224e+17,
|
| 4 |
+
"train_loss": 3.938131396484375,
|
| 5 |
+
"train_runtime": 22585.5788,
|
| 6 |
+
"train_samples": 960000,
|
| 7 |
+
"train_samples_per_second": 42.505,
|
| 8 |
+
"train_steps_per_second": 1.328
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,417 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 2000,
|
| 6 |
+
"global_step": 30000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 3.3333333333333335e-05,
|
| 13 |
+
"eval_accuracy": 0.016437927663734114,
|
| 14 |
+
"eval_loss": 10.702861785888672,
|
| 15 |
+
"eval_runtime": 53.5239,
|
| 16 |
+
"eval_samples_per_second": 93.416,
|
| 17 |
+
"eval_steps_per_second": 2.933,
|
| 18 |
+
"step": 1
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 6.666666666666667e-05,
|
| 22 |
+
"eval_accuracy": 0.049564027370478984,
|
| 23 |
+
"eval_loss": 10.533119201660156,
|
| 24 |
+
"eval_runtime": 53.3666,
|
| 25 |
+
"eval_samples_per_second": 93.692,
|
| 26 |
+
"eval_steps_per_second": 2.942,
|
| 27 |
+
"step": 2
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.00013333333333333334,
|
| 31 |
+
"eval_accuracy": 0.05325317693059629,
|
| 32 |
+
"eval_loss": 10.302159309387207,
|
| 33 |
+
"eval_runtime": 53.3789,
|
| 34 |
+
"eval_samples_per_second": 93.67,
|
| 35 |
+
"eval_steps_per_second": 2.941,
|
| 36 |
+
"step": 4
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 0.0002666666666666667,
|
| 40 |
+
"eval_accuracy": 0.053565591397849464,
|
| 41 |
+
"eval_loss": 10.023477554321289,
|
| 42 |
+
"eval_runtime": 53.6667,
|
| 43 |
+
"eval_samples_per_second": 93.168,
|
| 44 |
+
"eval_steps_per_second": 2.925,
|
| 45 |
+
"step": 8
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.0005333333333333334,
|
| 49 |
+
"eval_accuracy": 0.06345063538611925,
|
| 50 |
+
"eval_loss": 9.653569221496582,
|
| 51 |
+
"eval_runtime": 53.2342,
|
| 52 |
+
"eval_samples_per_second": 93.925,
|
| 53 |
+
"eval_steps_per_second": 2.949,
|
| 54 |
+
"step": 16
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.0010666666666666667,
|
| 58 |
+
"eval_accuracy": 0.07593020527859237,
|
| 59 |
+
"eval_loss": 9.028414726257324,
|
| 60 |
+
"eval_runtime": 53.4229,
|
| 61 |
+
"eval_samples_per_second": 93.593,
|
| 62 |
+
"eval_steps_per_second": 2.939,
|
| 63 |
+
"step": 32
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.0021333333333333334,
|
| 67 |
+
"eval_accuracy": 0.08323401759530792,
|
| 68 |
+
"eval_loss": 8.02489185333252,
|
| 69 |
+
"eval_runtime": 53.5928,
|
| 70 |
+
"eval_samples_per_second": 93.296,
|
| 71 |
+
"eval_steps_per_second": 2.929,
|
| 72 |
+
"step": 64
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.004266666666666667,
|
| 76 |
+
"eval_accuracy": 0.11289071358748778,
|
| 77 |
+
"eval_loss": 6.917232990264893,
|
| 78 |
+
"eval_runtime": 53.9308,
|
| 79 |
+
"eval_samples_per_second": 92.711,
|
| 80 |
+
"eval_steps_per_second": 2.911,
|
| 81 |
+
"step": 128
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.008533333333333334,
|
| 85 |
+
"eval_accuracy": 0.15581857282502443,
|
| 86 |
+
"eval_loss": 6.162940502166748,
|
| 87 |
+
"eval_runtime": 53.873,
|
| 88 |
+
"eval_samples_per_second": 92.811,
|
| 89 |
+
"eval_steps_per_second": 2.914,
|
| 90 |
+
"step": 256
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.017066666666666667,
|
| 94 |
+
"eval_accuracy": 0.18169071358748778,
|
| 95 |
+
"eval_loss": 5.580474376678467,
|
| 96 |
+
"eval_runtime": 53.993,
|
| 97 |
+
"eval_samples_per_second": 92.605,
|
| 98 |
+
"eval_steps_per_second": 2.908,
|
| 99 |
+
"step": 512
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.034133333333333335,
|
| 103 |
+
"eval_accuracy": 0.20284437927663734,
|
| 104 |
+
"eval_loss": 5.123527526855469,
|
| 105 |
+
"eval_runtime": 53.733,
|
| 106 |
+
"eval_samples_per_second": 93.053,
|
| 107 |
+
"eval_steps_per_second": 2.922,
|
| 108 |
+
"step": 1024
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.06666666666666667,
|
| 112 |
+
"grad_norm": 0.9611704349517822,
|
| 113 |
+
"learning_rate": 4.666666666666667e-05,
|
| 114 |
+
"loss": 5.4529,
|
| 115 |
+
"step": 2000
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.06666666666666667,
|
| 119 |
+
"eval_accuracy": 0.2264314760508309,
|
| 120 |
+
"eval_loss": 4.761301517486572,
|
| 121 |
+
"eval_runtime": 55.2965,
|
| 122 |
+
"eval_samples_per_second": 90.422,
|
| 123 |
+
"eval_steps_per_second": 2.839,
|
| 124 |
+
"step": 2000
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 0.06826666666666667,
|
| 128 |
+
"eval_accuracy": 0.22808172043010752,
|
| 129 |
+
"eval_loss": 4.7480597496032715,
|
| 130 |
+
"eval_runtime": 54.5009,
|
| 131 |
+
"eval_samples_per_second": 91.742,
|
| 132 |
+
"eval_steps_per_second": 2.881,
|
| 133 |
+
"step": 2048
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 0.13333333333333333,
|
| 137 |
+
"grad_norm": 0.9782880544662476,
|
| 138 |
+
"learning_rate": 4.3333333333333334e-05,
|
| 139 |
+
"loss": 4.5765,
|
| 140 |
+
"step": 4000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.13333333333333333,
|
| 144 |
+
"eval_accuracy": 0.2610080156402737,
|
| 145 |
+
"eval_loss": 4.412333011627197,
|
| 146 |
+
"eval_runtime": 55.0769,
|
| 147 |
+
"eval_samples_per_second": 90.782,
|
| 148 |
+
"eval_steps_per_second": 2.851,
|
| 149 |
+
"step": 4000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.13653333333333334,
|
| 153 |
+
"eval_accuracy": 0.2625071358748778,
|
| 154 |
+
"eval_loss": 4.404272556304932,
|
| 155 |
+
"eval_runtime": 54.9058,
|
| 156 |
+
"eval_samples_per_second": 91.065,
|
| 157 |
+
"eval_steps_per_second": 2.859,
|
| 158 |
+
"step": 4096
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.2,
|
| 162 |
+
"grad_norm": 1.0245925188064575,
|
| 163 |
+
"learning_rate": 4e-05,
|
| 164 |
+
"loss": 4.3252,
|
| 165 |
+
"step": 6000
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.2,
|
| 169 |
+
"eval_accuracy": 0.2826778103616813,
|
| 170 |
+
"eval_loss": 4.222066879272461,
|
| 171 |
+
"eval_runtime": 55.0772,
|
| 172 |
+
"eval_samples_per_second": 90.782,
|
| 173 |
+
"eval_steps_per_second": 2.851,
|
| 174 |
+
"step": 6000
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.26666666666666666,
|
| 178 |
+
"grad_norm": 1.9592970609664917,
|
| 179 |
+
"learning_rate": 3.6666666666666666e-05,
|
| 180 |
+
"loss": 4.146,
|
| 181 |
+
"step": 8000
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.26666666666666666,
|
| 185 |
+
"eval_accuracy": 0.3098099706744868,
|
| 186 |
+
"eval_loss": 4.0350446701049805,
|
| 187 |
+
"eval_runtime": 54.3944,
|
| 188 |
+
"eval_samples_per_second": 91.921,
|
| 189 |
+
"eval_steps_per_second": 2.886,
|
| 190 |
+
"step": 8000
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 0.2730666666666667,
|
| 194 |
+
"eval_accuracy": 0.31292649071358747,
|
| 195 |
+
"eval_loss": 4.013375759124756,
|
| 196 |
+
"eval_runtime": 54.918,
|
| 197 |
+
"eval_samples_per_second": 91.045,
|
| 198 |
+
"eval_steps_per_second": 2.859,
|
| 199 |
+
"step": 8192
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.3333333333333333,
|
| 203 |
+
"grad_norm": 1.1236926317214966,
|
| 204 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 205 |
+
"loss": 3.9652,
|
| 206 |
+
"step": 10000
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.3333333333333333,
|
| 210 |
+
"eval_accuracy": 0.3303505376344086,
|
| 211 |
+
"eval_loss": 3.885976552963257,
|
| 212 |
+
"eval_runtime": 54.6579,
|
| 213 |
+
"eval_samples_per_second": 91.478,
|
| 214 |
+
"eval_steps_per_second": 2.872,
|
| 215 |
+
"step": 10000
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 0.4,
|
| 219 |
+
"grad_norm": 1.4833592176437378,
|
| 220 |
+
"learning_rate": 3e-05,
|
| 221 |
+
"loss": 3.8441,
|
| 222 |
+
"step": 12000
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 0.4,
|
| 226 |
+
"eval_accuracy": 0.34179745845552295,
|
| 227 |
+
"eval_loss": 3.800518274307251,
|
| 228 |
+
"eval_runtime": 55.3363,
|
| 229 |
+
"eval_samples_per_second": 90.357,
|
| 230 |
+
"eval_steps_per_second": 2.837,
|
| 231 |
+
"step": 12000
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 0.4666666666666667,
|
| 235 |
+
"grad_norm": 1.413355827331543,
|
| 236 |
+
"learning_rate": 2.6666666666666667e-05,
|
| 237 |
+
"loss": 3.7739,
|
| 238 |
+
"step": 14000
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 0.4666666666666667,
|
| 242 |
+
"eval_accuracy": 0.3503352883675464,
|
| 243 |
+
"eval_loss": 3.7315235137939453,
|
| 244 |
+
"eval_runtime": 54.617,
|
| 245 |
+
"eval_samples_per_second": 91.547,
|
| 246 |
+
"eval_steps_per_second": 2.875,
|
| 247 |
+
"step": 14000
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.5333333333333333,
|
| 251 |
+
"grad_norm": 1.4252855777740479,
|
| 252 |
+
"learning_rate": 2.3333333333333336e-05,
|
| 253 |
+
"loss": 3.72,
|
| 254 |
+
"step": 16000
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.5333333333333333,
|
| 258 |
+
"eval_accuracy": 0.35525493646138806,
|
| 259 |
+
"eval_loss": 3.688023328781128,
|
| 260 |
+
"eval_runtime": 54.7137,
|
| 261 |
+
"eval_samples_per_second": 91.385,
|
| 262 |
+
"eval_steps_per_second": 2.869,
|
| 263 |
+
"step": 16000
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 0.5461333333333334,
|
| 267 |
+
"eval_accuracy": 0.356411339198436,
|
| 268 |
+
"eval_loss": 3.6776843070983887,
|
| 269 |
+
"eval_runtime": 55.1644,
|
| 270 |
+
"eval_samples_per_second": 90.638,
|
| 271 |
+
"eval_steps_per_second": 2.846,
|
| 272 |
+
"step": 16384
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.6,
|
| 276 |
+
"grad_norm": 1.4446940422058105,
|
| 277 |
+
"learning_rate": 2e-05,
|
| 278 |
+
"loss": 3.6718,
|
| 279 |
+
"step": 18000
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.6,
|
| 283 |
+
"eval_accuracy": 0.3592500488758553,
|
| 284 |
+
"eval_loss": 3.6533164978027344,
|
| 285 |
+
"eval_runtime": 53.9881,
|
| 286 |
+
"eval_samples_per_second": 92.613,
|
| 287 |
+
"eval_steps_per_second": 2.908,
|
| 288 |
+
"step": 18000
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 0.6666666666666666,
|
| 292 |
+
"grad_norm": 1.3672009706497192,
|
| 293 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 294 |
+
"loss": 3.6527,
|
| 295 |
+
"step": 20000
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 0.6666666666666666,
|
| 299 |
+
"eval_accuracy": 0.3633403714565005,
|
| 300 |
+
"eval_loss": 3.6212007999420166,
|
| 301 |
+
"eval_runtime": 53.6539,
|
| 302 |
+
"eval_samples_per_second": 93.19,
|
| 303 |
+
"eval_steps_per_second": 2.926,
|
| 304 |
+
"step": 20000
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.7333333333333333,
|
| 308 |
+
"grad_norm": 1.5204988718032837,
|
| 309 |
+
"learning_rate": 1.3333333333333333e-05,
|
| 310 |
+
"loss": 3.6201,
|
| 311 |
+
"step": 22000
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.7333333333333333,
|
| 315 |
+
"eval_accuracy": 0.36595581622678397,
|
| 316 |
+
"eval_loss": 3.598499298095703,
|
| 317 |
+
"eval_runtime": 54.0759,
|
| 318 |
+
"eval_samples_per_second": 92.463,
|
| 319 |
+
"eval_steps_per_second": 2.903,
|
| 320 |
+
"step": 22000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.8,
|
| 324 |
+
"grad_norm": 2.7004802227020264,
|
| 325 |
+
"learning_rate": 1e-05,
|
| 326 |
+
"loss": 3.593,
|
| 327 |
+
"step": 24000
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 0.8,
|
| 331 |
+
"eval_accuracy": 0.3679088954056696,
|
| 332 |
+
"eval_loss": 3.5819320678710938,
|
| 333 |
+
"eval_runtime": 53.8279,
|
| 334 |
+
"eval_samples_per_second": 92.889,
|
| 335 |
+
"eval_steps_per_second": 2.917,
|
| 336 |
+
"step": 24000
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 0.8666666666666667,
|
| 340 |
+
"grad_norm": 1.3153384923934937,
|
| 341 |
+
"learning_rate": 6.666666666666667e-06,
|
| 342 |
+
"loss": 3.5857,
|
| 343 |
+
"step": 26000
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 0.8666666666666667,
|
| 347 |
+
"eval_accuracy": 0.36973333333333336,
|
| 348 |
+
"eval_loss": 3.5683183670043945,
|
| 349 |
+
"eval_runtime": 54.3226,
|
| 350 |
+
"eval_samples_per_second": 92.043,
|
| 351 |
+
"eval_steps_per_second": 2.89,
|
| 352 |
+
"step": 26000
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.9333333333333333,
|
| 356 |
+
"grad_norm": 1.3367434740066528,
|
| 357 |
+
"learning_rate": 3.3333333333333333e-06,
|
| 358 |
+
"loss": 3.5801,
|
| 359 |
+
"step": 28000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.9333333333333333,
|
| 363 |
+
"eval_accuracy": 0.37106158357771263,
|
| 364 |
+
"eval_loss": 3.558154582977295,
|
| 365 |
+
"eval_runtime": 53.6794,
|
| 366 |
+
"eval_samples_per_second": 93.146,
|
| 367 |
+
"eval_steps_per_second": 2.925,
|
| 368 |
+
"step": 28000
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 1.0,
|
| 372 |
+
"grad_norm": 1.308415412902832,
|
| 373 |
+
"learning_rate": 0.0,
|
| 374 |
+
"loss": 3.5649,
|
| 375 |
+
"step": 30000
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 1.0,
|
| 379 |
+
"eval_accuracy": 0.3716248289345064,
|
| 380 |
+
"eval_loss": 3.5532469749450684,
|
| 381 |
+
"eval_runtime": 54.1176,
|
| 382 |
+
"eval_samples_per_second": 92.391,
|
| 383 |
+
"eval_steps_per_second": 2.901,
|
| 384 |
+
"step": 30000
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 1.0,
|
| 388 |
+
"step": 30000,
|
| 389 |
+
"total_flos": 2.6348013748224e+17,
|
| 390 |
+
"train_loss": 3.938131396484375,
|
| 391 |
+
"train_runtime": 22585.5788,
|
| 392 |
+
"train_samples_per_second": 42.505,
|
| 393 |
+
"train_steps_per_second": 1.328
|
| 394 |
+
}
|
| 395 |
+
],
|
| 396 |
+
"logging_steps": 2000,
|
| 397 |
+
"max_steps": 30000,
|
| 398 |
+
"num_input_tokens_seen": 0,
|
| 399 |
+
"num_train_epochs": 9223372036854775807,
|
| 400 |
+
"save_steps": 500,
|
| 401 |
+
"stateful_callbacks": {
|
| 402 |
+
"TrainerControl": {
|
| 403 |
+
"args": {
|
| 404 |
+
"should_epoch_stop": false,
|
| 405 |
+
"should_evaluate": false,
|
| 406 |
+
"should_log": false,
|
| 407 |
+
"should_save": true,
|
| 408 |
+
"should_training_stop": true
|
| 409 |
+
},
|
| 410 |
+
"attributes": {}
|
| 411 |
+
}
|
| 412 |
+
},
|
| 413 |
+
"total_flos": 2.6348013748224e+17,
|
| 414 |
+
"train_batch_size": 32,
|
| 415 |
+
"trial_name": null,
|
| 416 |
+
"trial_params": null
|
| 417 |
+
}
|