FINE-TUNED WITH MERGED DATASET
Browse files- README.md +14 -0
- all_results.json +8 -0
- runs/Mar03_09-41-24_a2727a3c73a1/events.out.tfevents.1709463658.a2727a3c73a1.593.1 +3 -0
- train_results.json +8 -0
- trainer_state.json +450 -0
README.md
CHANGED
|
@@ -3,6 +3,8 @@ license: apache-2.0
|
|
| 3 |
base_model: distilbert-base-uncased
|
| 4 |
tags:
|
| 5 |
- generated_from_trainer
|
|
|
|
|
|
|
| 6 |
model-index:
|
| 7 |
- name: finetuning-fake-reviews-detector-model
|
| 8 |
results: []
|
|
@@ -14,6 +16,18 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 14 |
# finetuning-fake-reviews-detector-model
|
| 15 |
|
| 16 |
This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the None dataset.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 17 |
|
| 18 |
## Model description
|
| 19 |
|
|
|
|
| 3 |
base_model: distilbert-base-uncased
|
| 4 |
tags:
|
| 5 |
- generated_from_trainer
|
| 6 |
+
metrics:
|
| 7 |
+
- accuracy
|
| 8 |
model-index:
|
| 9 |
- name: finetuning-fake-reviews-detector-model
|
| 10 |
results: []
|
|
|
|
| 16 |
# finetuning-fake-reviews-detector-model
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the None dataset.
|
| 19 |
+
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 0.4370
|
| 21 |
+
- Accuracy: 0.8844
|
| 22 |
+
- Weighted f1: 0.8821
|
| 23 |
+
- Micro f1: 0.8844
|
| 24 |
+
- Macro f1: 0.8151
|
| 25 |
+
- Weighted recall: 0.8844
|
| 26 |
+
- Micro recall: 0.8844
|
| 27 |
+
- Macro recall: 0.8032
|
| 28 |
+
- Weighted precision: 0.8808
|
| 29 |
+
- Micro precision: 0.8844
|
| 30 |
+
- Macro precision: 0.8293
|
| 31 |
|
| 32 |
## Model description
|
| 33 |
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"total_flos": 1.2565894692964032e+16,
|
| 4 |
+
"train_loss": 0.27979692445330456,
|
| 5 |
+
"train_runtime": 4649.1669,
|
| 6 |
+
"train_samples_per_second": 104.814,
|
| 7 |
+
"train_steps_per_second": 6.552
|
| 8 |
+
}
|
runs/Mar03_09-41-24_a2727a3c73a1/events.out.tfevents.1709463658.a2727a3c73a1.593.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:91ba59ddae8b0dcd15a5f5a2da464b49febd38a122734d9bd59981a60499c509
|
| 3 |
+
size 936
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"total_flos": 1.2565894692964032e+16,
|
| 4 |
+
"train_loss": 0.27979692445330456,
|
| 5 |
+
"train_runtime": 4649.1669,
|
| 6 |
+
"train_samples_per_second": 104.814,
|
| 7 |
+
"train_steps_per_second": 6.552
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,450 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 30460,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.08,
|
| 13 |
+
"grad_norm": 2.0977869033813477,
|
| 14 |
+
"learning_rate": 1.967170059093894e-05,
|
| 15 |
+
"loss": 0.4445,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.16,
|
| 20 |
+
"grad_norm": 4.845855236053467,
|
| 21 |
+
"learning_rate": 1.9343401181877874e-05,
|
| 22 |
+
"loss": 0.4078,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.25,
|
| 27 |
+
"grad_norm": 2.6214709281921387,
|
| 28 |
+
"learning_rate": 1.901510177281681e-05,
|
| 29 |
+
"loss": 0.3889,
|
| 30 |
+
"step": 1500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.33,
|
| 34 |
+
"grad_norm": 1.9115525484085083,
|
| 35 |
+
"learning_rate": 1.868680236375575e-05,
|
| 36 |
+
"loss": 0.378,
|
| 37 |
+
"step": 2000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.41,
|
| 41 |
+
"grad_norm": 1.6511852741241455,
|
| 42 |
+
"learning_rate": 1.8358502954694683e-05,
|
| 43 |
+
"loss": 0.3864,
|
| 44 |
+
"step": 2500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.49,
|
| 48 |
+
"grad_norm": 2.835732936859131,
|
| 49 |
+
"learning_rate": 1.803020354563362e-05,
|
| 50 |
+
"loss": 0.3795,
|
| 51 |
+
"step": 3000
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.57,
|
| 55 |
+
"grad_norm": 2.6627590656280518,
|
| 56 |
+
"learning_rate": 1.7701904136572555e-05,
|
| 57 |
+
"loss": 0.3709,
|
| 58 |
+
"step": 3500
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.66,
|
| 62 |
+
"grad_norm": 2.168017864227295,
|
| 63 |
+
"learning_rate": 1.7373604727511493e-05,
|
| 64 |
+
"loss": 0.3606,
|
| 65 |
+
"step": 4000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.74,
|
| 69 |
+
"grad_norm": 1.1796438694000244,
|
| 70 |
+
"learning_rate": 1.704530531845043e-05,
|
| 71 |
+
"loss": 0.375,
|
| 72 |
+
"step": 4500
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.82,
|
| 76 |
+
"grad_norm": 1.51750910282135,
|
| 77 |
+
"learning_rate": 1.6717005909389364e-05,
|
| 78 |
+
"loss": 0.3616,
|
| 79 |
+
"step": 5000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.9,
|
| 83 |
+
"grad_norm": 0.9447911381721497,
|
| 84 |
+
"learning_rate": 1.63887065003283e-05,
|
| 85 |
+
"loss": 0.3585,
|
| 86 |
+
"step": 5500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.98,
|
| 90 |
+
"grad_norm": 3.0596373081207275,
|
| 91 |
+
"learning_rate": 1.6060407091267236e-05,
|
| 92 |
+
"loss": 0.3501,
|
| 93 |
+
"step": 6000
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 1.07,
|
| 97 |
+
"grad_norm": 5.587700843811035,
|
| 98 |
+
"learning_rate": 1.5732107682206174e-05,
|
| 99 |
+
"loss": 0.3357,
|
| 100 |
+
"step": 6500
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 1.15,
|
| 104 |
+
"grad_norm": 2.992246389389038,
|
| 105 |
+
"learning_rate": 1.5403808273145108e-05,
|
| 106 |
+
"loss": 0.3269,
|
| 107 |
+
"step": 7000
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 1.23,
|
| 111 |
+
"grad_norm": 0.933838963508606,
|
| 112 |
+
"learning_rate": 1.5075508864084046e-05,
|
| 113 |
+
"loss": 0.3241,
|
| 114 |
+
"step": 7500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 1.31,
|
| 118 |
+
"grad_norm": 6.16735315322876,
|
| 119 |
+
"learning_rate": 1.4747209455022982e-05,
|
| 120 |
+
"loss": 0.3255,
|
| 121 |
+
"step": 8000
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 1.4,
|
| 125 |
+
"grad_norm": 6.236515522003174,
|
| 126 |
+
"learning_rate": 1.4418910045961918e-05,
|
| 127 |
+
"loss": 0.3412,
|
| 128 |
+
"step": 8500
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 1.48,
|
| 132 |
+
"grad_norm": 2.5656967163085938,
|
| 133 |
+
"learning_rate": 1.4090610636900854e-05,
|
| 134 |
+
"loss": 0.3349,
|
| 135 |
+
"step": 9000
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 1.56,
|
| 139 |
+
"grad_norm": 1.8035532236099243,
|
| 140 |
+
"learning_rate": 1.3762311227839791e-05,
|
| 141 |
+
"loss": 0.3133,
|
| 142 |
+
"step": 9500
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 1.64,
|
| 146 |
+
"grad_norm": 8.10788631439209,
|
| 147 |
+
"learning_rate": 1.3434011818778727e-05,
|
| 148 |
+
"loss": 0.3275,
|
| 149 |
+
"step": 10000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 1.72,
|
| 153 |
+
"grad_norm": 2.249290943145752,
|
| 154 |
+
"learning_rate": 1.3105712409717663e-05,
|
| 155 |
+
"loss": 0.3144,
|
| 156 |
+
"step": 10500
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 1.81,
|
| 160 |
+
"grad_norm": 3.0741331577301025,
|
| 161 |
+
"learning_rate": 1.27774130006566e-05,
|
| 162 |
+
"loss": 0.3198,
|
| 163 |
+
"step": 11000
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 1.89,
|
| 167 |
+
"grad_norm": 3.473233699798584,
|
| 168 |
+
"learning_rate": 1.2449113591595537e-05,
|
| 169 |
+
"loss": 0.31,
|
| 170 |
+
"step": 11500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 1.97,
|
| 174 |
+
"grad_norm": 3.777818441390991,
|
| 175 |
+
"learning_rate": 1.2120814182534473e-05,
|
| 176 |
+
"loss": 0.3272,
|
| 177 |
+
"step": 12000
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 2.05,
|
| 181 |
+
"grad_norm": 3.420518636703491,
|
| 182 |
+
"learning_rate": 1.1792514773473409e-05,
|
| 183 |
+
"loss": 0.2844,
|
| 184 |
+
"step": 12500
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 2.13,
|
| 188 |
+
"grad_norm": 0.9904893040657043,
|
| 189 |
+
"learning_rate": 1.1464215364412346e-05,
|
| 190 |
+
"loss": 0.285,
|
| 191 |
+
"step": 13000
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 2.22,
|
| 195 |
+
"grad_norm": 1.798154592514038,
|
| 196 |
+
"learning_rate": 1.1135915955351282e-05,
|
| 197 |
+
"loss": 0.269,
|
| 198 |
+
"step": 13500
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 2.3,
|
| 202 |
+
"grad_norm": 8.202375411987305,
|
| 203 |
+
"learning_rate": 1.0807616546290218e-05,
|
| 204 |
+
"loss": 0.2851,
|
| 205 |
+
"step": 14000
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 2.38,
|
| 209 |
+
"grad_norm": 6.387328147888184,
|
| 210 |
+
"learning_rate": 1.0479317137229154e-05,
|
| 211 |
+
"loss": 0.2713,
|
| 212 |
+
"step": 14500
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 2.46,
|
| 216 |
+
"grad_norm": 9.065672874450684,
|
| 217 |
+
"learning_rate": 1.0151017728168092e-05,
|
| 218 |
+
"loss": 0.2685,
|
| 219 |
+
"step": 15000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 2.54,
|
| 223 |
+
"grad_norm": 1.2352772951126099,
|
| 224 |
+
"learning_rate": 9.822718319107026e-06,
|
| 225 |
+
"loss": 0.2857,
|
| 226 |
+
"step": 15500
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 2.63,
|
| 230 |
+
"grad_norm": 8.483463287353516,
|
| 231 |
+
"learning_rate": 9.494418910045962e-06,
|
| 232 |
+
"loss": 0.2722,
|
| 233 |
+
"step": 16000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 2.71,
|
| 237 |
+
"grad_norm": 4.678401947021484,
|
| 238 |
+
"learning_rate": 9.1661195009849e-06,
|
| 239 |
+
"loss": 0.2654,
|
| 240 |
+
"step": 16500
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 2.79,
|
| 244 |
+
"grad_norm": 6.603830814361572,
|
| 245 |
+
"learning_rate": 8.837820091923836e-06,
|
| 246 |
+
"loss": 0.2748,
|
| 247 |
+
"step": 17000
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 2.87,
|
| 251 |
+
"grad_norm": 8.765490531921387,
|
| 252 |
+
"learning_rate": 8.509520682862772e-06,
|
| 253 |
+
"loss": 0.2742,
|
| 254 |
+
"step": 17500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 2.95,
|
| 258 |
+
"grad_norm": 0.8734021782875061,
|
| 259 |
+
"learning_rate": 8.181221273801707e-06,
|
| 260 |
+
"loss": 0.2897,
|
| 261 |
+
"step": 18000
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 3.04,
|
| 265 |
+
"grad_norm": 6.885056018829346,
|
| 266 |
+
"learning_rate": 7.852921864740645e-06,
|
| 267 |
+
"loss": 0.2477,
|
| 268 |
+
"step": 18500
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 3.12,
|
| 272 |
+
"grad_norm": 19.46624755859375,
|
| 273 |
+
"learning_rate": 7.52462245567958e-06,
|
| 274 |
+
"loss": 0.2134,
|
| 275 |
+
"step": 19000
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 3.2,
|
| 279 |
+
"grad_norm": 3.132298231124878,
|
| 280 |
+
"learning_rate": 7.196323046618516e-06,
|
| 281 |
+
"loss": 0.2377,
|
| 282 |
+
"step": 19500
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 3.28,
|
| 286 |
+
"grad_norm": 3.616292715072632,
|
| 287 |
+
"learning_rate": 6.868023637557453e-06,
|
| 288 |
+
"loss": 0.2314,
|
| 289 |
+
"step": 20000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 3.37,
|
| 293 |
+
"grad_norm": 36.205810546875,
|
| 294 |
+
"learning_rate": 6.539724228496389e-06,
|
| 295 |
+
"loss": 0.2253,
|
| 296 |
+
"step": 20500
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 3.45,
|
| 300 |
+
"grad_norm": 5.153477191925049,
|
| 301 |
+
"learning_rate": 6.211424819435326e-06,
|
| 302 |
+
"loss": 0.2289,
|
| 303 |
+
"step": 21000
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 3.53,
|
| 307 |
+
"grad_norm": 5.419785499572754,
|
| 308 |
+
"learning_rate": 5.883125410374262e-06,
|
| 309 |
+
"loss": 0.216,
|
| 310 |
+
"step": 21500
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 3.61,
|
| 314 |
+
"grad_norm": 1.2598270177841187,
|
| 315 |
+
"learning_rate": 5.554826001313198e-06,
|
| 316 |
+
"loss": 0.2354,
|
| 317 |
+
"step": 22000
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 3.69,
|
| 321 |
+
"grad_norm": 3.154592752456665,
|
| 322 |
+
"learning_rate": 5.226526592252134e-06,
|
| 323 |
+
"loss": 0.226,
|
| 324 |
+
"step": 22500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 3.78,
|
| 328 |
+
"grad_norm": 6.700149059295654,
|
| 329 |
+
"learning_rate": 4.89822718319107e-06,
|
| 330 |
+
"loss": 0.2322,
|
| 331 |
+
"step": 23000
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 3.86,
|
| 335 |
+
"grad_norm": 7.298892021179199,
|
| 336 |
+
"learning_rate": 4.569927774130007e-06,
|
| 337 |
+
"loss": 0.222,
|
| 338 |
+
"step": 23500
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 3.94,
|
| 342 |
+
"grad_norm": 3.5850930213928223,
|
| 343 |
+
"learning_rate": 4.241628365068943e-06,
|
| 344 |
+
"loss": 0.2348,
|
| 345 |
+
"step": 24000
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 4.02,
|
| 349 |
+
"grad_norm": 9.055290222167969,
|
| 350 |
+
"learning_rate": 3.91332895600788e-06,
|
| 351 |
+
"loss": 0.2222,
|
| 352 |
+
"step": 24500
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 4.1,
|
| 356 |
+
"grad_norm": 14.489812850952148,
|
| 357 |
+
"learning_rate": 3.585029546946816e-06,
|
| 358 |
+
"loss": 0.1874,
|
| 359 |
+
"step": 25000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 4.19,
|
| 363 |
+
"grad_norm": 5.70402717590332,
|
| 364 |
+
"learning_rate": 3.2567301378857517e-06,
|
| 365 |
+
"loss": 0.2021,
|
| 366 |
+
"step": 25500
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 4.27,
|
| 370 |
+
"grad_norm": 1.1003680229187012,
|
| 371 |
+
"learning_rate": 2.928430728824688e-06,
|
| 372 |
+
"loss": 0.1928,
|
| 373 |
+
"step": 26000
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 4.35,
|
| 377 |
+
"grad_norm": 20.254186630249023,
|
| 378 |
+
"learning_rate": 2.6001313197636244e-06,
|
| 379 |
+
"loss": 0.1994,
|
| 380 |
+
"step": 26500
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 4.43,
|
| 384 |
+
"grad_norm": 37.06382751464844,
|
| 385 |
+
"learning_rate": 2.2718319107025612e-06,
|
| 386 |
+
"loss": 0.1896,
|
| 387 |
+
"step": 27000
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 4.51,
|
| 391 |
+
"grad_norm": 30.246774673461914,
|
| 392 |
+
"learning_rate": 1.943532501641497e-06,
|
| 393 |
+
"loss": 0.1911,
|
| 394 |
+
"step": 27500
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 4.6,
|
| 398 |
+
"grad_norm": 1.3807177543640137,
|
| 399 |
+
"learning_rate": 1.6152330925804335e-06,
|
| 400 |
+
"loss": 0.1982,
|
| 401 |
+
"step": 28000
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 4.68,
|
| 405 |
+
"grad_norm": 18.111328125,
|
| 406 |
+
"learning_rate": 1.2869336835193697e-06,
|
| 407 |
+
"loss": 0.1931,
|
| 408 |
+
"step": 28500
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 4.76,
|
| 412 |
+
"grad_norm": 30.377147674560547,
|
| 413 |
+
"learning_rate": 9.58634274458306e-07,
|
| 414 |
+
"loss": 0.1897,
|
| 415 |
+
"step": 29000
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 4.84,
|
| 419 |
+
"grad_norm": 7.666069507598877,
|
| 420 |
+
"learning_rate": 6.303348653972423e-07,
|
| 421 |
+
"loss": 0.1876,
|
| 422 |
+
"step": 29500
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 4.92,
|
| 426 |
+
"grad_norm": 15.851324081420898,
|
| 427 |
+
"learning_rate": 3.020354563361786e-07,
|
| 428 |
+
"loss": 0.1757,
|
| 429 |
+
"step": 30000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 5.0,
|
| 433 |
+
"step": 30460,
|
| 434 |
+
"total_flos": 1.2565894692964032e+16,
|
| 435 |
+
"train_loss": 0.27979692445330456,
|
| 436 |
+
"train_runtime": 4649.1669,
|
| 437 |
+
"train_samples_per_second": 104.814,
|
| 438 |
+
"train_steps_per_second": 6.552
|
| 439 |
+
}
|
| 440 |
+
],
|
| 441 |
+
"logging_steps": 500,
|
| 442 |
+
"max_steps": 30460,
|
| 443 |
+
"num_input_tokens_seen": 0,
|
| 444 |
+
"num_train_epochs": 5,
|
| 445 |
+
"save_steps": 500,
|
| 446 |
+
"total_flos": 1.2565894692964032e+16,
|
| 447 |
+
"train_batch_size": 16,
|
| 448 |
+
"trial_name": null,
|
| 449 |
+
"trial_params": null
|
| 450 |
+
}
|