End of training
Browse files- README.md +14 -0
- adapter_model.safetensors +1 -1
- all_results.json +5 -5
- train_results.json +5 -5
- trainer_state.json +393 -13
README.md
CHANGED
|
@@ -15,6 +15,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
# Mistral-7B-v0.1_caselaw
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the None dataset.
|
|
|
|
|
|
|
| 18 |
|
| 19 |
## Model description
|
| 20 |
|
|
@@ -49,6 +51,18 @@ The following hyperparameters were used during training:
|
|
| 49 |
|
| 50 |
### Training results
|
| 51 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 52 |
|
| 53 |
|
| 54 |
### Framework versions
|
|
|
|
| 15 |
# Mistral-7B-v0.1_caselaw
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the None dataset.
|
| 18 |
+
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 1.2884
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
|
|
|
| 51 |
|
| 52 |
### Training results
|
| 53 |
|
| 54 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 55 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 56 |
+
| 1.309 | 0.19 | 50 | 1.3631 |
|
| 57 |
+
| 1.2966 | 0.39 | 100 | 1.3340 |
|
| 58 |
+
| 1.2914 | 0.58 | 150 | 1.3168 |
|
| 59 |
+
| 1.298 | 0.77 | 200 | 1.3039 |
|
| 60 |
+
| 1.2678 | 0.96 | 250 | 1.2991 |
|
| 61 |
+
| 1.216 | 1.16 | 300 | 1.3008 |
|
| 62 |
+
| 1.2467 | 1.35 | 350 | 1.2945 |
|
| 63 |
+
| 1.223 | 1.54 | 400 | 1.2940 |
|
| 64 |
+
| 1.2 | 1.74 | 450 | 1.2924 |
|
| 65 |
+
| 1.2406 | 1.93 | 500 | 1.2884 |
|
| 66 |
|
| 67 |
|
| 68 |
### Framework versions
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 83946192
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f8e5ecbf1b1c5f51d16a600bc209e8495214ac7d5f2a492bb2e9feae430537fc
|
| 3 |
size 83946192
|
all_results.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss": 1.
|
| 4 |
-
"train_runtime":
|
| 5 |
-
"train_samples_per_second":
|
| 6 |
-
"train_steps_per_second": 0.
|
| 7 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 2.0,
|
| 3 |
+
"train_loss": 1.2557248572124937,
|
| 4 |
+
"train_runtime": 2763.6306,
|
| 5 |
+
"train_samples_per_second": 12.007,
|
| 6 |
+
"train_steps_per_second": 0.187
|
| 7 |
}
|
train_results.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss": 1.
|
| 4 |
-
"train_runtime":
|
| 5 |
-
"train_samples_per_second":
|
| 6 |
-
"train_steps_per_second": 0.
|
| 7 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 2.0,
|
| 3 |
+
"train_loss": 1.2557248572124937,
|
| 4 |
+
"train_runtime": 2763.6306,
|
| 5 |
+
"train_samples_per_second": 12.007,
|
| 6 |
+
"train_steps_per_second": 0.187
|
| 7 |
}
|
trainer_state.json
CHANGED
|
@@ -1,35 +1,415 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch": 1.
|
| 5 |
"eval_steps": 50,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch":
|
| 13 |
"learning_rate": 5e-05,
|
| 14 |
-
"loss": 1.
|
| 15 |
"step": 10
|
| 16 |
},
|
| 17 |
{
|
| 18 |
-
"epoch":
|
| 19 |
-
"
|
| 20 |
-
"
|
| 21 |
-
"
|
| 22 |
-
|
| 23 |
-
|
| 24 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 25 |
}
|
| 26 |
],
|
| 27 |
"logging_steps": 10,
|
| 28 |
-
"max_steps":
|
| 29 |
"num_input_tokens_seen": 0,
|
| 30 |
"num_train_epochs": 2,
|
| 31 |
"save_steps": 100,
|
| 32 |
-
"total_flos":
|
| 33 |
"train_batch_size": 4,
|
| 34 |
"trial_name": null,
|
| 35 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.9980713596914175,
|
| 5 |
"eval_steps": 50,
|
| 6 |
+
"global_step": 518,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.04,
|
| 13 |
"learning_rate": 5e-05,
|
| 14 |
+
"loss": 1.5247,
|
| 15 |
"step": 10
|
| 16 |
},
|
| 17 |
{
|
| 18 |
+
"epoch": 0.08,
|
| 19 |
+
"learning_rate": 5e-05,
|
| 20 |
+
"loss": 1.4114,
|
| 21 |
+
"step": 20
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.12,
|
| 25 |
+
"learning_rate": 5e-05,
|
| 26 |
+
"loss": 1.3203,
|
| 27 |
+
"step": 30
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.15,
|
| 31 |
+
"learning_rate": 5e-05,
|
| 32 |
+
"loss": 1.3566,
|
| 33 |
+
"step": 40
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.19,
|
| 37 |
+
"learning_rate": 5e-05,
|
| 38 |
+
"loss": 1.309,
|
| 39 |
+
"step": 50
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.19,
|
| 43 |
+
"eval_loss": 1.3630590438842773,
|
| 44 |
+
"eval_runtime": 26.2031,
|
| 45 |
+
"eval_samples_per_second": 33.355,
|
| 46 |
+
"eval_steps_per_second": 2.099,
|
| 47 |
+
"step": 50
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 0.23,
|
| 51 |
+
"learning_rate": 5e-05,
|
| 52 |
+
"loss": 1.2856,
|
| 53 |
+
"step": 60
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.27,
|
| 57 |
+
"learning_rate": 5e-05,
|
| 58 |
+
"loss": 1.3524,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.31,
|
| 63 |
+
"learning_rate": 5e-05,
|
| 64 |
+
"loss": 1.317,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.35,
|
| 69 |
+
"learning_rate": 5e-05,
|
| 70 |
+
"loss": 1.324,
|
| 71 |
+
"step": 90
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 0.39,
|
| 75 |
+
"learning_rate": 5e-05,
|
| 76 |
+
"loss": 1.2966,
|
| 77 |
+
"step": 100
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 0.39,
|
| 81 |
+
"eval_loss": 1.3339532613754272,
|
| 82 |
+
"eval_runtime": 25.949,
|
| 83 |
+
"eval_samples_per_second": 33.681,
|
| 84 |
+
"eval_steps_per_second": 2.12,
|
| 85 |
+
"step": 100
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 0.42,
|
| 89 |
+
"learning_rate": 5e-05,
|
| 90 |
+
"loss": 1.3073,
|
| 91 |
+
"step": 110
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.46,
|
| 95 |
+
"learning_rate": 5e-05,
|
| 96 |
+
"loss": 1.3162,
|
| 97 |
+
"step": 120
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.5,
|
| 101 |
+
"learning_rate": 5e-05,
|
| 102 |
+
"loss": 1.3299,
|
| 103 |
+
"step": 130
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 0.54,
|
| 107 |
+
"learning_rate": 5e-05,
|
| 108 |
+
"loss": 1.3271,
|
| 109 |
+
"step": 140
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.58,
|
| 113 |
+
"learning_rate": 5e-05,
|
| 114 |
+
"loss": 1.2914,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.58,
|
| 119 |
+
"eval_loss": 1.3168174028396606,
|
| 120 |
+
"eval_runtime": 26.0065,
|
| 121 |
+
"eval_samples_per_second": 33.607,
|
| 122 |
+
"eval_steps_per_second": 2.115,
|
| 123 |
+
"step": 150
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.62,
|
| 127 |
+
"learning_rate": 5e-05,
|
| 128 |
+
"loss": 1.2526,
|
| 129 |
+
"step": 160
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.66,
|
| 133 |
+
"learning_rate": 5e-05,
|
| 134 |
+
"loss": 1.2924,
|
| 135 |
+
"step": 170
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.69,
|
| 139 |
+
"learning_rate": 5e-05,
|
| 140 |
+
"loss": 1.3303,
|
| 141 |
+
"step": 180
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.73,
|
| 145 |
+
"learning_rate": 5e-05,
|
| 146 |
+
"loss": 1.3173,
|
| 147 |
+
"step": 190
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 0.77,
|
| 151 |
+
"learning_rate": 5e-05,
|
| 152 |
+
"loss": 1.298,
|
| 153 |
+
"step": 200
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.77,
|
| 157 |
+
"eval_loss": 1.303916096687317,
|
| 158 |
+
"eval_runtime": 26.2236,
|
| 159 |
+
"eval_samples_per_second": 33.329,
|
| 160 |
+
"eval_steps_per_second": 2.097,
|
| 161 |
+
"step": 200
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 0.81,
|
| 165 |
+
"learning_rate": 5e-05,
|
| 166 |
+
"loss": 1.2402,
|
| 167 |
+
"step": 210
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 0.85,
|
| 171 |
+
"learning_rate": 5e-05,
|
| 172 |
+
"loss": 1.2768,
|
| 173 |
+
"step": 220
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 0.89,
|
| 177 |
+
"learning_rate": 5e-05,
|
| 178 |
+
"loss": 1.2929,
|
| 179 |
+
"step": 230
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 0.93,
|
| 183 |
+
"learning_rate": 5e-05,
|
| 184 |
+
"loss": 1.2744,
|
| 185 |
+
"step": 240
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.96,
|
| 189 |
+
"learning_rate": 5e-05,
|
| 190 |
+
"loss": 1.2678,
|
| 191 |
+
"step": 250
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.96,
|
| 195 |
+
"eval_loss": 1.2991052865982056,
|
| 196 |
+
"eval_runtime": 26.0876,
|
| 197 |
+
"eval_samples_per_second": 33.502,
|
| 198 |
+
"eval_steps_per_second": 2.108,
|
| 199 |
+
"step": 250
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 1.0,
|
| 203 |
+
"learning_rate": 5e-05,
|
| 204 |
+
"loss": 1.2506,
|
| 205 |
+
"step": 260
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 1.04,
|
| 209 |
+
"learning_rate": 5e-05,
|
| 210 |
+
"loss": 1.1717,
|
| 211 |
+
"step": 270
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 1.08,
|
| 215 |
+
"learning_rate": 5e-05,
|
| 216 |
+
"loss": 1.2022,
|
| 217 |
+
"step": 280
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 1.12,
|
| 221 |
+
"learning_rate": 5e-05,
|
| 222 |
+
"loss": 1.2237,
|
| 223 |
+
"step": 290
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 1.16,
|
| 227 |
+
"learning_rate": 5e-05,
|
| 228 |
+
"loss": 1.216,
|
| 229 |
+
"step": 300
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 1.16,
|
| 233 |
+
"eval_loss": 1.3007503747940063,
|
| 234 |
+
"eval_runtime": 28.1383,
|
| 235 |
+
"eval_samples_per_second": 31.061,
|
| 236 |
+
"eval_steps_per_second": 1.955,
|
| 237 |
+
"step": 300
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 1.2,
|
| 241 |
+
"learning_rate": 5e-05,
|
| 242 |
+
"loss": 1.2264,
|
| 243 |
+
"step": 310
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 1.23,
|
| 247 |
+
"learning_rate": 5e-05,
|
| 248 |
+
"loss": 1.1461,
|
| 249 |
+
"step": 320
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 1.27,
|
| 253 |
+
"learning_rate": 5e-05,
|
| 254 |
+
"loss": 1.1855,
|
| 255 |
+
"step": 330
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 1.31,
|
| 259 |
+
"learning_rate": 5e-05,
|
| 260 |
+
"loss": 1.2014,
|
| 261 |
+
"step": 340
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 1.35,
|
| 265 |
+
"learning_rate": 5e-05,
|
| 266 |
+
"loss": 1.2467,
|
| 267 |
+
"step": 350
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 1.35,
|
| 271 |
+
"eval_loss": 1.2945101261138916,
|
| 272 |
+
"eval_runtime": 25.9821,
|
| 273 |
+
"eval_samples_per_second": 33.639,
|
| 274 |
+
"eval_steps_per_second": 2.117,
|
| 275 |
+
"step": 350
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 1.39,
|
| 279 |
+
"learning_rate": 5e-05,
|
| 280 |
+
"loss": 1.2136,
|
| 281 |
+
"step": 360
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 1.43,
|
| 285 |
+
"learning_rate": 5e-05,
|
| 286 |
+
"loss": 1.1727,
|
| 287 |
+
"step": 370
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 1.47,
|
| 291 |
+
"learning_rate": 5e-05,
|
| 292 |
+
"loss": 1.192,
|
| 293 |
+
"step": 380
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 1.5,
|
| 297 |
+
"learning_rate": 5e-05,
|
| 298 |
+
"loss": 1.1963,
|
| 299 |
+
"step": 390
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 1.54,
|
| 303 |
+
"learning_rate": 5e-05,
|
| 304 |
+
"loss": 1.223,
|
| 305 |
+
"step": 400
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 1.54,
|
| 309 |
+
"eval_loss": 1.293986201286316,
|
| 310 |
+
"eval_runtime": 26.8556,
|
| 311 |
+
"eval_samples_per_second": 32.544,
|
| 312 |
+
"eval_steps_per_second": 2.048,
|
| 313 |
+
"step": 400
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 1.58,
|
| 317 |
+
"learning_rate": 5e-05,
|
| 318 |
+
"loss": 1.2076,
|
| 319 |
+
"step": 410
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 1.62,
|
| 323 |
+
"learning_rate": 5e-05,
|
| 324 |
+
"loss": 1.1373,
|
| 325 |
+
"step": 420
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 1.66,
|
| 329 |
+
"learning_rate": 5e-05,
|
| 330 |
+
"loss": 1.224,
|
| 331 |
+
"step": 430
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 1.7,
|
| 335 |
+
"learning_rate": 5e-05,
|
| 336 |
+
"loss": 1.1841,
|
| 337 |
+
"step": 440
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 1.74,
|
| 341 |
+
"learning_rate": 5e-05,
|
| 342 |
+
"loss": 1.2,
|
| 343 |
+
"step": 450
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 1.74,
|
| 347 |
+
"eval_loss": 1.2924402952194214,
|
| 348 |
+
"eval_runtime": 29.7855,
|
| 349 |
+
"eval_samples_per_second": 29.343,
|
| 350 |
+
"eval_steps_per_second": 1.847,
|
| 351 |
+
"step": 450
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 1.77,
|
| 355 |
+
"learning_rate": 5e-05,
|
| 356 |
+
"loss": 1.1896,
|
| 357 |
+
"step": 460
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 1.81,
|
| 361 |
+
"learning_rate": 5e-05,
|
| 362 |
+
"loss": 1.1476,
|
| 363 |
+
"step": 470
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 1.85,
|
| 367 |
+
"learning_rate": 5e-05,
|
| 368 |
+
"loss": 1.1986,
|
| 369 |
+
"step": 480
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 1.89,
|
| 373 |
+
"learning_rate": 5e-05,
|
| 374 |
+
"loss": 1.2185,
|
| 375 |
+
"step": 490
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 1.93,
|
| 379 |
+
"learning_rate": 5e-05,
|
| 380 |
+
"loss": 1.2406,
|
| 381 |
+
"step": 500
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 1.93,
|
| 385 |
+
"eval_loss": 1.2884066104888916,
|
| 386 |
+
"eval_runtime": 26.2052,
|
| 387 |
+
"eval_samples_per_second": 33.352,
|
| 388 |
+
"eval_steps_per_second": 2.099,
|
| 389 |
+
"step": 500
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 1.97,
|
| 393 |
+
"learning_rate": 5e-05,
|
| 394 |
+
"loss": 1.1962,
|
| 395 |
+
"step": 510
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 2.0,
|
| 399 |
+
"step": 518,
|
| 400 |
+
"total_flos": 5.99394721244119e+17,
|
| 401 |
+
"train_loss": 1.2557248572124937,
|
| 402 |
+
"train_runtime": 2763.6306,
|
| 403 |
+
"train_samples_per_second": 12.007,
|
| 404 |
+
"train_steps_per_second": 0.187
|
| 405 |
}
|
| 406 |
],
|
| 407 |
"logging_steps": 10,
|
| 408 |
+
"max_steps": 518,
|
| 409 |
"num_input_tokens_seen": 0,
|
| 410 |
"num_train_epochs": 2,
|
| 411 |
"save_steps": 100,
|
| 412 |
+
"total_flos": 5.99394721244119e+17,
|
| 413 |
"train_batch_size": 4,
|
| 414 |
"trial_name": null,
|
| 415 |
"trial_params": null
|