Model save
Browse files- README.md +8 -13
- all_results.json +6 -11
- model.safetensors +1 -1
- runs/Dec09_09-43-44_2caa3c6c30c3/events.out.tfevents.1733737433.2caa3c6c30c3.23.2 +3 -0
- train_results.json +6 -6
- trainer_state.json +451 -106
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -18,8 +18,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on an unknown dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
-
- Loss: 0.
|
| 22 |
-
- Accuracy: 0.
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
@@ -44,23 +44,18 @@ The following hyperparameters were used during training:
|
|
| 44 |
- seed: 42
|
| 45 |
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
| 46 |
- lr_scheduler_type: linear
|
| 47 |
-
- num_epochs:
|
| 48 |
- mixed_precision_training: Native AMP
|
| 49 |
|
| 50 |
### Training results
|
| 51 |
|
| 52 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 53 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 54 |
-
| 0.
|
| 55 |
-
| 0.
|
| 56 |
-
| 0.
|
| 57 |
-
| 0.
|
| 58 |
-
| 0.
|
| 59 |
-
| 0.0221 | 6.0 | 384 | 0.7708 | 0.8883 |
|
| 60 |
-
| 0.0202 | 7.0 | 448 | 0.7649 | 0.8911 |
|
| 61 |
-
| 0.0274 | 8.0 | 512 | 0.7099 | 0.9162 |
|
| 62 |
-
| 0.0507 | 9.0 | 576 | 0.7499 | 0.9162 |
|
| 63 |
-
| 0.0653 | 10.0 | 640 | 0.7514 | 0.9162 |
|
| 64 |
|
| 65 |
|
| 66 |
### Framework versions
|
|
|
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on an unknown dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.8713
|
| 22 |
+
- Accuracy: 0.9050
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
|
|
| 44 |
- seed: 42
|
| 45 |
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
| 46 |
- lr_scheduler_type: linear
|
| 47 |
+
- num_epochs: 5
|
| 48 |
- mixed_precision_training: Native AMP
|
| 49 |
|
| 50 |
### Training results
|
| 51 |
|
| 52 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 53 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 54 |
+
| 0.2055 | 1.0 | 64 | 1.0229 | 0.8436 |
|
| 55 |
+
| 0.1685 | 2.0 | 128 | 0.8520 | 0.8855 |
|
| 56 |
+
| 0.0168 | 3.0 | 192 | 0.8259 | 0.8994 |
|
| 57 |
+
| 0.054 | 4.0 | 256 | 0.8865 | 0.8911 |
|
| 58 |
+
| 0.0236 | 5.0 | 320 | 0.8713 | 0.9050 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 59 |
|
| 60 |
|
| 61 |
### Framework versions
|
all_results.json
CHANGED
|
@@ -1,13 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"
|
| 4 |
-
"
|
| 5 |
-
"
|
| 6 |
-
"
|
| 7 |
-
"
|
| 8 |
-
"total_flos": 6.287051649950024e+17,
|
| 9 |
-
"train_loss": 0.7197832907550037,
|
| 10 |
-
"train_runtime": 268.6443,
|
| 11 |
-
"train_samples_per_second": 30.196,
|
| 12 |
-
"train_steps_per_second": 0.953
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 1.571762912487506e+18,
|
| 4 |
+
"train_loss": 0.0904881817754358,
|
| 5 |
+
"train_runtime": 704.9996,
|
| 6 |
+
"train_samples_per_second": 28.766,
|
| 7 |
+
"train_steps_per_second": 0.908
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 8 |
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 343273192
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:832edbfb27793dddda33fbe268905cefc2184cc20f75f24d2671a8ae2d070ee8
|
| 3 |
size 343273192
|
runs/Dec09_09-43-44_2caa3c6c30c3/events.out.tfevents.1733737433.2caa3c6c30c3.23.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b67aaa5bd113f5f1b6a1062998b057bb8b53825a18a3b71e3d24d94a8dddd2db
|
| 3 |
+
size 15129
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 1.571762912487506e+18,
|
| 4 |
+
"train_loss": 0.0904881817754358,
|
| 5 |
+
"train_runtime": 704.9996,
|
| 6 |
+
"train_samples_per_second": 28.766,
|
| 7 |
+
"train_steps_per_second": 0.908
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,221 +1,566 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "finetuned-bangladeshi-traditional-food/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
-
"eval_steps":
|
| 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": 0.15625,
|
| 13 |
-
"grad_norm":
|
| 14 |
-
"learning_rate": 0.
|
| 15 |
-
"loss":
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.3125,
|
| 20 |
-
"grad_norm":
|
| 21 |
-
"learning_rate": 0.
|
| 22 |
-
"loss":
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.46875,
|
| 27 |
-
"grad_norm":
|
| 28 |
-
"learning_rate": 0.
|
| 29 |
-
"loss":
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.625,
|
| 34 |
-
"grad_norm":
|
| 35 |
-
"learning_rate": 0.
|
| 36 |
-
"loss":
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.78125,
|
| 41 |
-
"grad_norm":
|
| 42 |
-
"learning_rate": 0.
|
| 43 |
-
"loss":
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.9375,
|
| 48 |
-
"grad_norm":
|
| 49 |
-
"learning_rate": 0.
|
| 50 |
-
"loss":
|
| 51 |
"step": 60
|
| 52 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 53 |
{
|
| 54 |
"epoch": 1.09375,
|
| 55 |
-
"grad_norm":
|
| 56 |
-
"learning_rate": 0.
|
| 57 |
-
"loss": 0.
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 1.25,
|
| 62 |
-
"grad_norm":
|
| 63 |
-
"learning_rate": 0.
|
| 64 |
-
"loss": 0.
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 1.40625,
|
| 69 |
-
"grad_norm":
|
| 70 |
-
"learning_rate": 0.
|
| 71 |
-
"loss": 0.
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 1.5625,
|
| 76 |
-
"grad_norm":
|
| 77 |
-
"learning_rate": 0.
|
| 78 |
-
"loss": 0.
|
| 79 |
-
"step": 100
|
| 80 |
-
},
|
| 81 |
-
{
|
| 82 |
-
"epoch": 1.5625,
|
| 83 |
-
"eval_accuracy": 0.9022346368715084,
|
| 84 |
-
"eval_loss": 0.612750232219696,
|
| 85 |
-
"eval_runtime": 6.1858,
|
| 86 |
-
"eval_samples_per_second": 57.874,
|
| 87 |
-
"eval_steps_per_second": 3.718,
|
| 88 |
"step": 100
|
| 89 |
},
|
| 90 |
{
|
| 91 |
"epoch": 1.71875,
|
| 92 |
-
"grad_norm":
|
| 93 |
-
"learning_rate": 0.
|
| 94 |
-
"loss": 0.
|
| 95 |
"step": 110
|
| 96 |
},
|
| 97 |
{
|
| 98 |
"epoch": 1.875,
|
| 99 |
-
"grad_norm":
|
| 100 |
-
"learning_rate": 0.
|
| 101 |
-
"loss": 0.
|
| 102 |
"step": 120
|
| 103 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 104 |
{
|
| 105 |
"epoch": 2.03125,
|
| 106 |
-
"grad_norm":
|
| 107 |
-
"learning_rate":
|
| 108 |
-
"loss": 0.
|
| 109 |
"step": 130
|
| 110 |
},
|
| 111 |
{
|
| 112 |
"epoch": 2.1875,
|
| 113 |
-
"grad_norm":
|
| 114 |
-
"learning_rate":
|
| 115 |
-
"loss": 0.
|
| 116 |
"step": 140
|
| 117 |
},
|
| 118 |
{
|
| 119 |
"epoch": 2.34375,
|
| 120 |
-
"grad_norm":
|
| 121 |
-
"learning_rate":
|
| 122 |
-
"loss": 0.
|
| 123 |
"step": 150
|
| 124 |
},
|
| 125 |
{
|
| 126 |
"epoch": 2.5,
|
| 127 |
-
"grad_norm":
|
| 128 |
-
"learning_rate":
|
| 129 |
-
"loss": 0.
|
| 130 |
"step": 160
|
| 131 |
},
|
| 132 |
{
|
| 133 |
"epoch": 2.65625,
|
| 134 |
-
"grad_norm":
|
| 135 |
-
"learning_rate":
|
| 136 |
-
"loss": 0.
|
| 137 |
"step": 170
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"epoch": 2.8125,
|
| 141 |
-
"grad_norm":
|
| 142 |
-
"learning_rate":
|
| 143 |
-
"loss": 0.
|
| 144 |
"step": 180
|
| 145 |
},
|
| 146 |
{
|
| 147 |
"epoch": 2.96875,
|
| 148 |
-
"grad_norm":
|
| 149 |
-
"learning_rate":
|
| 150 |
-
"loss": 0.
|
| 151 |
"step": 190
|
| 152 |
},
|
| 153 |
{
|
| 154 |
-
"epoch": 3.
|
| 155 |
-
"
|
| 156 |
-
"
|
| 157 |
-
"
|
| 158 |
-
"
|
|
|
|
|
|
|
| 159 |
},
|
| 160 |
{
|
| 161 |
"epoch": 3.125,
|
| 162 |
-
"
|
| 163 |
-
"
|
| 164 |
-
"
|
| 165 |
-
"eval_samples_per_second": 58.67,
|
| 166 |
-
"eval_steps_per_second": 3.769,
|
| 167 |
"step": 200
|
| 168 |
},
|
| 169 |
{
|
| 170 |
"epoch": 3.28125,
|
| 171 |
-
"grad_norm":
|
| 172 |
-
"learning_rate":
|
| 173 |
-
"loss": 0.
|
| 174 |
"step": 210
|
| 175 |
},
|
| 176 |
{
|
| 177 |
"epoch": 3.4375,
|
| 178 |
-
"grad_norm":
|
| 179 |
-
"learning_rate":
|
| 180 |
-
"loss": 0.
|
| 181 |
"step": 220
|
| 182 |
},
|
| 183 |
{
|
| 184 |
"epoch": 3.59375,
|
| 185 |
-
"grad_norm":
|
| 186 |
-
"learning_rate":
|
| 187 |
-
"loss": 0.
|
| 188 |
"step": 230
|
| 189 |
},
|
| 190 |
{
|
| 191 |
"epoch": 3.75,
|
| 192 |
-
"grad_norm":
|
| 193 |
-
"learning_rate":
|
| 194 |
-
"loss": 0.
|
| 195 |
"step": 240
|
| 196 |
},
|
| 197 |
{
|
| 198 |
"epoch": 3.90625,
|
| 199 |
-
"grad_norm":
|
| 200 |
-
"learning_rate":
|
| 201 |
-
"loss": 0.
|
| 202 |
"step": 250
|
| 203 |
},
|
| 204 |
{
|
| 205 |
"epoch": 4.0,
|
| 206 |
-
"
|
| 207 |
-
"
|
| 208 |
-
"
|
| 209 |
-
"
|
| 210 |
-
"
|
| 211 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 212 |
}
|
| 213 |
],
|
| 214 |
"logging_steps": 10,
|
| 215 |
-
"max_steps":
|
| 216 |
"num_input_tokens_seen": 0,
|
| 217 |
-
"num_train_epochs":
|
| 218 |
-
"save_steps":
|
| 219 |
"stateful_callbacks": {
|
| 220 |
"TrainerControl": {
|
| 221 |
"args": {
|
|
@@ -228,7 +573,7 @@
|
|
| 228 |
"attributes": {}
|
| 229 |
}
|
| 230 |
},
|
| 231 |
-
"total_flos":
|
| 232 |
"train_batch_size": 32,
|
| 233 |
"trial_name": null,
|
| 234 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.6337994337081909,
|
| 3 |
+
"best_model_checkpoint": "finetuned-bangladeshi-traditional-food/checkpoint-320",
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 640,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.15625,
|
| 13 |
+
"grad_norm": 260322.40625,
|
| 14 |
+
"learning_rate": 0.000196875,
|
| 15 |
+
"loss": 0.1664,
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.3125,
|
| 20 |
+
"grad_norm": 9524.54296875,
|
| 21 |
+
"learning_rate": 0.00019375000000000002,
|
| 22 |
+
"loss": 0.132,
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.46875,
|
| 27 |
+
"grad_norm": 232655.25,
|
| 28 |
+
"learning_rate": 0.000190625,
|
| 29 |
+
"loss": 0.2052,
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.625,
|
| 34 |
+
"grad_norm": 2797.121337890625,
|
| 35 |
+
"learning_rate": 0.0001875,
|
| 36 |
+
"loss": 0.1608,
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.78125,
|
| 41 |
+
"grad_norm": 94306.1328125,
|
| 42 |
+
"learning_rate": 0.000184375,
|
| 43 |
+
"loss": 0.0879,
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.9375,
|
| 48 |
+
"grad_norm": 94746.8828125,
|
| 49 |
+
"learning_rate": 0.00018125000000000001,
|
| 50 |
+
"loss": 0.2159,
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 1.0,
|
| 55 |
+
"eval_accuracy": 0.840782122905028,
|
| 56 |
+
"eval_loss": 0.6425307393074036,
|
| 57 |
+
"eval_runtime": 6.1943,
|
| 58 |
+
"eval_samples_per_second": 57.795,
|
| 59 |
+
"eval_steps_per_second": 3.713,
|
| 60 |
+
"step": 64
|
| 61 |
+
},
|
| 62 |
{
|
| 63 |
"epoch": 1.09375,
|
| 64 |
+
"grad_norm": 224044.515625,
|
| 65 |
+
"learning_rate": 0.000178125,
|
| 66 |
+
"loss": 0.1128,
|
| 67 |
"step": 70
|
| 68 |
},
|
| 69 |
{
|
| 70 |
"epoch": 1.25,
|
| 71 |
+
"grad_norm": 179013.890625,
|
| 72 |
+
"learning_rate": 0.000175,
|
| 73 |
+
"loss": 0.1238,
|
| 74 |
"step": 80
|
| 75 |
},
|
| 76 |
{
|
| 77 |
"epoch": 1.40625,
|
| 78 |
+
"grad_norm": 322362.1875,
|
| 79 |
+
"learning_rate": 0.00017187500000000002,
|
| 80 |
+
"loss": 0.0767,
|
| 81 |
"step": 90
|
| 82 |
},
|
| 83 |
{
|
| 84 |
"epoch": 1.5625,
|
| 85 |
+
"grad_norm": 486709.90625,
|
| 86 |
+
"learning_rate": 0.00016875,
|
| 87 |
+
"loss": 0.1934,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 88 |
"step": 100
|
| 89 |
},
|
| 90 |
{
|
| 91 |
"epoch": 1.71875,
|
| 92 |
+
"grad_norm": 339304.5,
|
| 93 |
+
"learning_rate": 0.000165625,
|
| 94 |
+
"loss": 0.1545,
|
| 95 |
"step": 110
|
| 96 |
},
|
| 97 |
{
|
| 98 |
"epoch": 1.875,
|
| 99 |
+
"grad_norm": 400540.96875,
|
| 100 |
+
"learning_rate": 0.00016250000000000002,
|
| 101 |
+
"loss": 0.1968,
|
| 102 |
"step": 120
|
| 103 |
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 2.0,
|
| 106 |
+
"eval_accuracy": 0.8547486033519553,
|
| 107 |
+
"eval_loss": 0.6727585792541504,
|
| 108 |
+
"eval_runtime": 6.1785,
|
| 109 |
+
"eval_samples_per_second": 57.943,
|
| 110 |
+
"eval_steps_per_second": 3.723,
|
| 111 |
+
"step": 128
|
| 112 |
+
},
|
| 113 |
{
|
| 114 |
"epoch": 2.03125,
|
| 115 |
+
"grad_norm": 402626.6875,
|
| 116 |
+
"learning_rate": 0.000159375,
|
| 117 |
+
"loss": 0.1303,
|
| 118 |
"step": 130
|
| 119 |
},
|
| 120 |
{
|
| 121 |
"epoch": 2.1875,
|
| 122 |
+
"grad_norm": 84834.3515625,
|
| 123 |
+
"learning_rate": 0.00015625,
|
| 124 |
+
"loss": 0.1479,
|
| 125 |
"step": 140
|
| 126 |
},
|
| 127 |
{
|
| 128 |
"epoch": 2.34375,
|
| 129 |
+
"grad_norm": 120292.5703125,
|
| 130 |
+
"learning_rate": 0.000153125,
|
| 131 |
+
"loss": 0.1937,
|
| 132 |
"step": 150
|
| 133 |
},
|
| 134 |
{
|
| 135 |
"epoch": 2.5,
|
| 136 |
+
"grad_norm": 20186.486328125,
|
| 137 |
+
"learning_rate": 0.00015000000000000001,
|
| 138 |
+
"loss": 0.0804,
|
| 139 |
"step": 160
|
| 140 |
},
|
| 141 |
{
|
| 142 |
"epoch": 2.65625,
|
| 143 |
+
"grad_norm": 173780.65625,
|
| 144 |
+
"learning_rate": 0.000146875,
|
| 145 |
+
"loss": 0.0843,
|
| 146 |
"step": 170
|
| 147 |
},
|
| 148 |
{
|
| 149 |
"epoch": 2.8125,
|
| 150 |
+
"grad_norm": 46256.0546875,
|
| 151 |
+
"learning_rate": 0.00014375,
|
| 152 |
+
"loss": 0.0831,
|
| 153 |
"step": 180
|
| 154 |
},
|
| 155 |
{
|
| 156 |
"epoch": 2.96875,
|
| 157 |
+
"grad_norm": 165973.46875,
|
| 158 |
+
"learning_rate": 0.00014062500000000002,
|
| 159 |
+
"loss": 0.0686,
|
| 160 |
"step": 190
|
| 161 |
},
|
| 162 |
{
|
| 163 |
+
"epoch": 3.0,
|
| 164 |
+
"eval_accuracy": 0.8631284916201117,
|
| 165 |
+
"eval_loss": 0.71826171875,
|
| 166 |
+
"eval_runtime": 6.475,
|
| 167 |
+
"eval_samples_per_second": 55.29,
|
| 168 |
+
"eval_steps_per_second": 3.552,
|
| 169 |
+
"step": 192
|
| 170 |
},
|
| 171 |
{
|
| 172 |
"epoch": 3.125,
|
| 173 |
+
"grad_norm": 811.771240234375,
|
| 174 |
+
"learning_rate": 0.0001375,
|
| 175 |
+
"loss": 0.0824,
|
|
|
|
|
|
|
| 176 |
"step": 200
|
| 177 |
},
|
| 178 |
{
|
| 179 |
"epoch": 3.28125,
|
| 180 |
+
"grad_norm": 421833.125,
|
| 181 |
+
"learning_rate": 0.000134375,
|
| 182 |
+
"loss": 0.1451,
|
| 183 |
"step": 210
|
| 184 |
},
|
| 185 |
{
|
| 186 |
"epoch": 3.4375,
|
| 187 |
+
"grad_norm": 347591.78125,
|
| 188 |
+
"learning_rate": 0.00013125000000000002,
|
| 189 |
+
"loss": 0.1193,
|
| 190 |
"step": 220
|
| 191 |
},
|
| 192 |
{
|
| 193 |
"epoch": 3.59375,
|
| 194 |
+
"grad_norm": 150345.671875,
|
| 195 |
+
"learning_rate": 0.000128125,
|
| 196 |
+
"loss": 0.1245,
|
| 197 |
"step": 230
|
| 198 |
},
|
| 199 |
{
|
| 200 |
"epoch": 3.75,
|
| 201 |
+
"grad_norm": 370.7461853027344,
|
| 202 |
+
"learning_rate": 0.000125,
|
| 203 |
+
"loss": 0.0934,
|
| 204 |
"step": 240
|
| 205 |
},
|
| 206 |
{
|
| 207 |
"epoch": 3.90625,
|
| 208 |
+
"grad_norm": 583907.0,
|
| 209 |
+
"learning_rate": 0.00012187500000000001,
|
| 210 |
+
"loss": 0.0975,
|
| 211 |
"step": 250
|
| 212 |
},
|
| 213 |
{
|
| 214 |
"epoch": 4.0,
|
| 215 |
+
"eval_accuracy": 0.8463687150837989,
|
| 216 |
+
"eval_loss": 0.8238765597343445,
|
| 217 |
+
"eval_runtime": 6.2123,
|
| 218 |
+
"eval_samples_per_second": 57.628,
|
| 219 |
+
"eval_steps_per_second": 3.702,
|
| 220 |
+
"step": 256
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 4.0625,
|
| 224 |
+
"grad_norm": 75910.390625,
|
| 225 |
+
"learning_rate": 0.00011875,
|
| 226 |
+
"loss": 0.0952,
|
| 227 |
+
"step": 260
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 4.21875,
|
| 231 |
+
"grad_norm": 483136.6875,
|
| 232 |
+
"learning_rate": 0.000115625,
|
| 233 |
+
"loss": 0.0867,
|
| 234 |
+
"step": 270
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 4.375,
|
| 238 |
+
"grad_norm": 4663.18310546875,
|
| 239 |
+
"learning_rate": 0.00011250000000000001,
|
| 240 |
+
"loss": 0.0599,
|
| 241 |
+
"step": 280
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 4.53125,
|
| 245 |
+
"grad_norm": 1170.08251953125,
|
| 246 |
+
"learning_rate": 0.000109375,
|
| 247 |
+
"loss": 0.0671,
|
| 248 |
+
"step": 290
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 4.6875,
|
| 252 |
+
"grad_norm": 286.7016296386719,
|
| 253 |
+
"learning_rate": 0.00010625000000000001,
|
| 254 |
+
"loss": 0.1158,
|
| 255 |
+
"step": 300
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 4.84375,
|
| 259 |
+
"grad_norm": 237.0006561279297,
|
| 260 |
+
"learning_rate": 0.000103125,
|
| 261 |
+
"loss": 0.0821,
|
| 262 |
+
"step": 310
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 5.0,
|
| 266 |
+
"grad_norm": 137.74867248535156,
|
| 267 |
+
"learning_rate": 0.0001,
|
| 268 |
+
"loss": 0.0399,
|
| 269 |
+
"step": 320
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 5.0,
|
| 273 |
+
"eval_accuracy": 0.888268156424581,
|
| 274 |
+
"eval_loss": 0.6337994337081909,
|
| 275 |
+
"eval_runtime": 6.2754,
|
| 276 |
+
"eval_samples_per_second": 57.048,
|
| 277 |
+
"eval_steps_per_second": 3.665,
|
| 278 |
+
"step": 320
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 5.15625,
|
| 282 |
+
"grad_norm": 1214.9869384765625,
|
| 283 |
+
"learning_rate": 9.687500000000001e-05,
|
| 284 |
+
"loss": 0.1064,
|
| 285 |
+
"step": 330
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 5.3125,
|
| 289 |
+
"grad_norm": 534309.6875,
|
| 290 |
+
"learning_rate": 9.375e-05,
|
| 291 |
+
"loss": 0.0826,
|
| 292 |
+
"step": 340
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 5.46875,
|
| 296 |
+
"grad_norm": 2779.36669921875,
|
| 297 |
+
"learning_rate": 9.062500000000001e-05,
|
| 298 |
+
"loss": 0.0524,
|
| 299 |
+
"step": 350
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 5.625,
|
| 303 |
+
"grad_norm": 2379.431640625,
|
| 304 |
+
"learning_rate": 8.75e-05,
|
| 305 |
+
"loss": 0.059,
|
| 306 |
+
"step": 360
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 5.78125,
|
| 310 |
+
"grad_norm": 567636.125,
|
| 311 |
+
"learning_rate": 8.4375e-05,
|
| 312 |
+
"loss": 0.1303,
|
| 313 |
+
"step": 370
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 5.9375,
|
| 317 |
+
"grad_norm": 582.1480712890625,
|
| 318 |
+
"learning_rate": 8.125000000000001e-05,
|
| 319 |
+
"loss": 0.0221,
|
| 320 |
+
"step": 380
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 6.0,
|
| 324 |
+
"eval_accuracy": 0.888268156424581,
|
| 325 |
+
"eval_loss": 0.7707696557044983,
|
| 326 |
+
"eval_runtime": 6.201,
|
| 327 |
+
"eval_samples_per_second": 57.732,
|
| 328 |
+
"eval_steps_per_second": 3.709,
|
| 329 |
+
"step": 384
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 6.09375,
|
| 333 |
+
"grad_norm": 60.556556701660156,
|
| 334 |
+
"learning_rate": 7.8125e-05,
|
| 335 |
+
"loss": 0.0341,
|
| 336 |
+
"step": 390
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 6.25,
|
| 340 |
+
"grad_norm": 59.626529693603516,
|
| 341 |
+
"learning_rate": 7.500000000000001e-05,
|
| 342 |
+
"loss": 0.047,
|
| 343 |
+
"step": 400
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 6.40625,
|
| 347 |
+
"grad_norm": 303419.21875,
|
| 348 |
+
"learning_rate": 7.1875e-05,
|
| 349 |
+
"loss": 0.065,
|
| 350 |
+
"step": 410
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 6.5625,
|
| 354 |
+
"grad_norm": 90.21700286865234,
|
| 355 |
+
"learning_rate": 6.875e-05,
|
| 356 |
+
"loss": 0.0541,
|
| 357 |
+
"step": 420
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 6.71875,
|
| 361 |
+
"grad_norm": 801652.9375,
|
| 362 |
+
"learning_rate": 6.562500000000001e-05,
|
| 363 |
+
"loss": 0.1373,
|
| 364 |
+
"step": 430
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 6.875,
|
| 368 |
+
"grad_norm": 281023.09375,
|
| 369 |
+
"learning_rate": 6.25e-05,
|
| 370 |
+
"loss": 0.0202,
|
| 371 |
+
"step": 440
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 7.0,
|
| 375 |
+
"eval_accuracy": 0.8910614525139665,
|
| 376 |
+
"eval_loss": 0.7648816704750061,
|
| 377 |
+
"eval_runtime": 6.2429,
|
| 378 |
+
"eval_samples_per_second": 57.345,
|
| 379 |
+
"eval_steps_per_second": 3.684,
|
| 380 |
+
"step": 448
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 7.03125,
|
| 384 |
+
"grad_norm": 43.21717834472656,
|
| 385 |
+
"learning_rate": 5.9375e-05,
|
| 386 |
+
"loss": 0.0798,
|
| 387 |
+
"step": 450
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 7.1875,
|
| 391 |
+
"grad_norm": 622758.6875,
|
| 392 |
+
"learning_rate": 5.6250000000000005e-05,
|
| 393 |
+
"loss": 0.0585,
|
| 394 |
+
"step": 460
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 7.34375,
|
| 398 |
+
"grad_norm": 4335.7919921875,
|
| 399 |
+
"learning_rate": 5.3125000000000004e-05,
|
| 400 |
+
"loss": 0.0967,
|
| 401 |
+
"step": 470
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 7.5,
|
| 405 |
+
"grad_norm": 56.46971130371094,
|
| 406 |
+
"learning_rate": 5e-05,
|
| 407 |
+
"loss": 0.0412,
|
| 408 |
+
"step": 480
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 7.65625,
|
| 412 |
+
"grad_norm": 26.425151824951172,
|
| 413 |
+
"learning_rate": 4.6875e-05,
|
| 414 |
+
"loss": 0.0834,
|
| 415 |
+
"step": 490
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 7.8125,
|
| 419 |
+
"grad_norm": 2260.0234375,
|
| 420 |
+
"learning_rate": 4.375e-05,
|
| 421 |
+
"loss": 0.0237,
|
| 422 |
+
"step": 500
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 7.96875,
|
| 426 |
+
"grad_norm": 55191.5859375,
|
| 427 |
+
"learning_rate": 4.0625000000000005e-05,
|
| 428 |
+
"loss": 0.0274,
|
| 429 |
+
"step": 510
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 8.0,
|
| 433 |
+
"eval_accuracy": 0.9162011173184358,
|
| 434 |
+
"eval_loss": 0.7099100351333618,
|
| 435 |
+
"eval_runtime": 6.3314,
|
| 436 |
+
"eval_samples_per_second": 56.543,
|
| 437 |
+
"eval_steps_per_second": 3.633,
|
| 438 |
+
"step": 512
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 8.125,
|
| 442 |
+
"grad_norm": 34213.54296875,
|
| 443 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 444 |
+
"loss": 0.0295,
|
| 445 |
+
"step": 520
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 8.28125,
|
| 449 |
+
"grad_norm": 35.443115234375,
|
| 450 |
+
"learning_rate": 3.4375e-05,
|
| 451 |
+
"loss": 0.0814,
|
| 452 |
+
"step": 530
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 8.4375,
|
| 456 |
+
"grad_norm": 141960.859375,
|
| 457 |
+
"learning_rate": 3.125e-05,
|
| 458 |
+
"loss": 0.0377,
|
| 459 |
+
"step": 540
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 8.59375,
|
| 463 |
+
"grad_norm": 40429.1171875,
|
| 464 |
+
"learning_rate": 2.8125000000000003e-05,
|
| 465 |
+
"loss": 0.0481,
|
| 466 |
+
"step": 550
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 8.75,
|
| 470 |
+
"grad_norm": 11.297977447509766,
|
| 471 |
+
"learning_rate": 2.5e-05,
|
| 472 |
+
"loss": 0.0573,
|
| 473 |
+
"step": 560
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 8.90625,
|
| 477 |
+
"grad_norm": 12.592033386230469,
|
| 478 |
+
"learning_rate": 2.1875e-05,
|
| 479 |
+
"loss": 0.0507,
|
| 480 |
+
"step": 570
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 9.0,
|
| 484 |
+
"eval_accuracy": 0.9162011173184358,
|
| 485 |
+
"eval_loss": 0.7499300837516785,
|
| 486 |
+
"eval_runtime": 6.5756,
|
| 487 |
+
"eval_samples_per_second": 54.444,
|
| 488 |
+
"eval_steps_per_second": 3.498,
|
| 489 |
+
"step": 576
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 9.0625,
|
| 493 |
+
"grad_norm": 38.957550048828125,
|
| 494 |
+
"learning_rate": 1.8750000000000002e-05,
|
| 495 |
+
"loss": 0.0137,
|
| 496 |
+
"step": 580
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 9.21875,
|
| 500 |
+
"grad_norm": 37808.00390625,
|
| 501 |
+
"learning_rate": 1.5625e-05,
|
| 502 |
+
"loss": 0.1018,
|
| 503 |
+
"step": 590
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 9.375,
|
| 507 |
+
"grad_norm": 35257.2890625,
|
| 508 |
+
"learning_rate": 1.25e-05,
|
| 509 |
+
"loss": 0.0305,
|
| 510 |
+
"step": 600
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 9.53125,
|
| 514 |
+
"grad_norm": 14.713729858398438,
|
| 515 |
+
"learning_rate": 9.375000000000001e-06,
|
| 516 |
+
"loss": 0.0538,
|
| 517 |
+
"step": 610
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 9.6875,
|
| 521 |
+
"grad_norm": 123944.28125,
|
| 522 |
+
"learning_rate": 6.25e-06,
|
| 523 |
+
"loss": 0.0499,
|
| 524 |
+
"step": 620
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 9.84375,
|
| 528 |
+
"grad_norm": 199116.203125,
|
| 529 |
+
"learning_rate": 3.125e-06,
|
| 530 |
+
"loss": 0.1264,
|
| 531 |
+
"step": 630
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 10.0,
|
| 535 |
+
"grad_norm": 156690.1875,
|
| 536 |
+
"learning_rate": 0.0,
|
| 537 |
+
"loss": 0.0653,
|
| 538 |
+
"step": 640
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 10.0,
|
| 542 |
+
"eval_accuracy": 0.9162011173184358,
|
| 543 |
+
"eval_loss": 0.7513992786407471,
|
| 544 |
+
"eval_runtime": 6.3086,
|
| 545 |
+
"eval_samples_per_second": 56.748,
|
| 546 |
+
"eval_steps_per_second": 3.646,
|
| 547 |
+
"step": 640
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 10.0,
|
| 551 |
+
"step": 640,
|
| 552 |
+
"total_flos": 1.571762912487506e+18,
|
| 553 |
+
"train_loss": 0.0904881817754358,
|
| 554 |
+
"train_runtime": 704.9996,
|
| 555 |
+
"train_samples_per_second": 28.766,
|
| 556 |
+
"train_steps_per_second": 0.908
|
| 557 |
}
|
| 558 |
],
|
| 559 |
"logging_steps": 10,
|
| 560 |
+
"max_steps": 640,
|
| 561 |
"num_input_tokens_seen": 0,
|
| 562 |
+
"num_train_epochs": 10,
|
| 563 |
+
"save_steps": 500,
|
| 564 |
"stateful_callbacks": {
|
| 565 |
"TrainerControl": {
|
| 566 |
"args": {
|
|
|
|
| 573 |
"attributes": {}
|
| 574 |
}
|
| 575 |
},
|
| 576 |
+
"total_flos": 1.571762912487506e+18,
|
| 577 |
"train_batch_size": 32,
|
| 578 |
"trial_name": null,
|
| 579 |
"trial_params": null
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5304
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3be451d57d4ab8404b63e638e193d508db59c7a217ad8c217a377d4a6fcca80a
|
| 3 |
size 5304
|