Commit
•
e2fc4dc
1
Parent(s):
d5ec288
End of training
Browse files- README.md +13 -0
- trainer_state.json +493 -0
README.md
CHANGED
|
@@ -7,6 +7,7 @@ metrics:
|
|
| 7 |
model-index:
|
| 8 |
- name: llama-2-13b-reward-oasst1
|
| 9 |
results: []
|
|
|
|
| 10 |
---
|
| 11 |
|
| 12 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -33,6 +34,17 @@ More information needed
|
|
| 33 |
|
| 34 |
## Training procedure
|
| 35 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 36 |
### Training hyperparameters
|
| 37 |
|
| 38 |
The following hyperparameters were used during training:
|
|
@@ -66,6 +78,7 @@ The following hyperparameters were used during training:
|
|
| 66 |
|
| 67 |
### Framework versions
|
| 68 |
|
|
|
|
| 69 |
- Transformers 4.32.0.dev0
|
| 70 |
- Pytorch 2.0.1+cu118
|
| 71 |
- Datasets 2.14.0
|
|
|
|
| 7 |
model-index:
|
| 8 |
- name: llama-2-13b-reward-oasst1
|
| 9 |
results: []
|
| 10 |
+
library_name: peft
|
| 11 |
---
|
| 12 |
|
| 13 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 34 |
|
| 35 |
## Training procedure
|
| 36 |
|
| 37 |
+
|
| 38 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 39 |
+
- load_in_8bit: False
|
| 40 |
+
- load_in_4bit: True
|
| 41 |
+
- llm_int8_threshold: 6.0
|
| 42 |
+
- llm_int8_skip_modules: None
|
| 43 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 44 |
+
- llm_int8_has_fp16_weight: False
|
| 45 |
+
- bnb_4bit_quant_type: nf4
|
| 46 |
+
- bnb_4bit_use_double_quant: False
|
| 47 |
+
- bnb_4bit_compute_dtype: float16
|
| 48 |
### Training hyperparameters
|
| 49 |
|
| 50 |
The following hyperparameters were used during training:
|
|
|
|
| 78 |
|
| 79 |
### Framework versions
|
| 80 |
|
| 81 |
+
- PEFT 0.5.0.dev0
|
| 82 |
- Transformers 4.32.0.dev0
|
| 83 |
- Pytorch 2.0.1+cu118
|
| 84 |
- Datasets 2.14.0
|
trainer_state.json
ADDED
|
@@ -0,0 +1,493 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9998333611064822,
|
| 5 |
+
"global_step": 3000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.02,
|
| 12 |
+
"learning_rate": 1.968e-05,
|
| 13 |
+
"loss": 0.8156,
|
| 14 |
+
"step": 50
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.03,
|
| 18 |
+
"learning_rate": 1.934666666666667e-05,
|
| 19 |
+
"loss": 0.7572,
|
| 20 |
+
"step": 100
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.05,
|
| 24 |
+
"learning_rate": 1.902e-05,
|
| 25 |
+
"loss": 0.743,
|
| 26 |
+
"step": 150
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.07,
|
| 30 |
+
"learning_rate": 1.868666666666667e-05,
|
| 31 |
+
"loss": 0.647,
|
| 32 |
+
"step": 200
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.08,
|
| 36 |
+
"learning_rate": 1.8353333333333333e-05,
|
| 37 |
+
"loss": 0.5602,
|
| 38 |
+
"step": 250
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.08,
|
| 42 |
+
"eval_accuracy": 0.7387820512820513,
|
| 43 |
+
"eval_loss": 0.5436390042304993,
|
| 44 |
+
"eval_runtime": 503.4806,
|
| 45 |
+
"eval_samples_per_second": 1.239,
|
| 46 |
+
"eval_steps_per_second": 1.239,
|
| 47 |
+
"step": 250
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 0.1,
|
| 51 |
+
"learning_rate": 1.802e-05,
|
| 52 |
+
"loss": 0.5444,
|
| 53 |
+
"step": 300
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.12,
|
| 57 |
+
"learning_rate": 1.7686666666666668e-05,
|
| 58 |
+
"loss": 0.6125,
|
| 59 |
+
"step": 350
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.13,
|
| 63 |
+
"learning_rate": 1.7353333333333335e-05,
|
| 64 |
+
"loss": 0.6015,
|
| 65 |
+
"step": 400
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.15,
|
| 69 |
+
"learning_rate": 1.702e-05,
|
| 70 |
+
"loss": 0.4994,
|
| 71 |
+
"step": 450
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 0.17,
|
| 75 |
+
"learning_rate": 1.6686666666666667e-05,
|
| 76 |
+
"loss": 0.6166,
|
| 77 |
+
"step": 500
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 0.17,
|
| 81 |
+
"eval_accuracy": 0.7467948717948718,
|
| 82 |
+
"eval_loss": 0.5339850187301636,
|
| 83 |
+
"eval_runtime": 503.4677,
|
| 84 |
+
"eval_samples_per_second": 1.239,
|
| 85 |
+
"eval_steps_per_second": 1.239,
|
| 86 |
+
"step": 500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.18,
|
| 90 |
+
"learning_rate": 1.6353333333333335e-05,
|
| 91 |
+
"loss": 0.4642,
|
| 92 |
+
"step": 550
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.2,
|
| 96 |
+
"learning_rate": 1.6020000000000002e-05,
|
| 97 |
+
"loss": 0.5976,
|
| 98 |
+
"step": 600
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.22,
|
| 102 |
+
"learning_rate": 1.5686666666666667e-05,
|
| 103 |
+
"loss": 0.6385,
|
| 104 |
+
"step": 650
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.23,
|
| 108 |
+
"learning_rate": 1.5353333333333334e-05,
|
| 109 |
+
"loss": 0.5903,
|
| 110 |
+
"step": 700
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.25,
|
| 114 |
+
"learning_rate": 1.5020000000000002e-05,
|
| 115 |
+
"loss": 0.6545,
|
| 116 |
+
"step": 750
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.25,
|
| 120 |
+
"eval_accuracy": 0.7644230769230769,
|
| 121 |
+
"eval_loss": 0.48992109298706055,
|
| 122 |
+
"eval_runtime": 503.2366,
|
| 123 |
+
"eval_samples_per_second": 1.24,
|
| 124 |
+
"eval_steps_per_second": 1.24,
|
| 125 |
+
"step": 750
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 0.27,
|
| 129 |
+
"learning_rate": 1.4686666666666666e-05,
|
| 130 |
+
"loss": 0.5017,
|
| 131 |
+
"step": 800
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 0.28,
|
| 135 |
+
"learning_rate": 1.4353333333333335e-05,
|
| 136 |
+
"loss": 0.5559,
|
| 137 |
+
"step": 850
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.3,
|
| 141 |
+
"learning_rate": 1.402e-05,
|
| 142 |
+
"loss": 0.5902,
|
| 143 |
+
"step": 900
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.32,
|
| 147 |
+
"learning_rate": 1.3686666666666669e-05,
|
| 148 |
+
"loss": 0.6273,
|
| 149 |
+
"step": 950
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.33,
|
| 153 |
+
"learning_rate": 1.3353333333333333e-05,
|
| 154 |
+
"loss": 0.5635,
|
| 155 |
+
"step": 1000
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 0.33,
|
| 159 |
+
"eval_accuracy": 0.7532051282051282,
|
| 160 |
+
"eval_loss": 0.487714022397995,
|
| 161 |
+
"eval_runtime": 502.9177,
|
| 162 |
+
"eval_samples_per_second": 1.241,
|
| 163 |
+
"eval_steps_per_second": 1.241,
|
| 164 |
+
"step": 1000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.35,
|
| 168 |
+
"learning_rate": 1.302e-05,
|
| 169 |
+
"loss": 0.5657,
|
| 170 |
+
"step": 1050
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.37,
|
| 174 |
+
"learning_rate": 1.2686666666666667e-05,
|
| 175 |
+
"loss": 0.4788,
|
| 176 |
+
"step": 1100
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.38,
|
| 180 |
+
"learning_rate": 1.2353333333333334e-05,
|
| 181 |
+
"loss": 0.6025,
|
| 182 |
+
"step": 1150
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.4,
|
| 186 |
+
"learning_rate": 1.202e-05,
|
| 187 |
+
"loss": 0.5794,
|
| 188 |
+
"step": 1200
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.42,
|
| 192 |
+
"learning_rate": 1.1686666666666668e-05,
|
| 193 |
+
"loss": 0.5933,
|
| 194 |
+
"step": 1250
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.42,
|
| 198 |
+
"eval_accuracy": 0.7660256410256411,
|
| 199 |
+
"eval_loss": 0.49303892254829407,
|
| 200 |
+
"eval_runtime": 503.4551,
|
| 201 |
+
"eval_samples_per_second": 1.239,
|
| 202 |
+
"eval_steps_per_second": 1.239,
|
| 203 |
+
"step": 1250
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 0.43,
|
| 207 |
+
"learning_rate": 1.1353333333333334e-05,
|
| 208 |
+
"loss": 0.5455,
|
| 209 |
+
"step": 1300
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.45,
|
| 213 |
+
"learning_rate": 1.1020000000000001e-05,
|
| 214 |
+
"loss": 0.6124,
|
| 215 |
+
"step": 1350
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 0.47,
|
| 219 |
+
"learning_rate": 1.0686666666666667e-05,
|
| 220 |
+
"loss": 0.6029,
|
| 221 |
+
"step": 1400
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 0.48,
|
| 225 |
+
"learning_rate": 1.0353333333333335e-05,
|
| 226 |
+
"loss": 0.5182,
|
| 227 |
+
"step": 1450
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.5,
|
| 231 |
+
"learning_rate": 1.002e-05,
|
| 232 |
+
"loss": 0.5758,
|
| 233 |
+
"step": 1500
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.5,
|
| 237 |
+
"eval_accuracy": 0.7740384615384616,
|
| 238 |
+
"eval_loss": 0.48509156703948975,
|
| 239 |
+
"eval_runtime": 503.7874,
|
| 240 |
+
"eval_samples_per_second": 1.239,
|
| 241 |
+
"eval_steps_per_second": 1.239,
|
| 242 |
+
"step": 1500
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.52,
|
| 246 |
+
"learning_rate": 9.686666666666668e-06,
|
| 247 |
+
"loss": 0.3444,
|
| 248 |
+
"step": 1550
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.53,
|
| 252 |
+
"learning_rate": 9.353333333333334e-06,
|
| 253 |
+
"loss": 0.5743,
|
| 254 |
+
"step": 1600
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.55,
|
| 258 |
+
"learning_rate": 9.020000000000002e-06,
|
| 259 |
+
"loss": 0.4628,
|
| 260 |
+
"step": 1650
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.57,
|
| 264 |
+
"learning_rate": 8.686666666666668e-06,
|
| 265 |
+
"loss": 0.5054,
|
| 266 |
+
"step": 1700
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 0.58,
|
| 270 |
+
"learning_rate": 8.353333333333335e-06,
|
| 271 |
+
"loss": 0.5212,
|
| 272 |
+
"step": 1750
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.58,
|
| 276 |
+
"eval_accuracy": 0.7788461538461539,
|
| 277 |
+
"eval_loss": 0.5020816326141357,
|
| 278 |
+
"eval_runtime": 503.024,
|
| 279 |
+
"eval_samples_per_second": 1.24,
|
| 280 |
+
"eval_steps_per_second": 1.24,
|
| 281 |
+
"step": 1750
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 0.6,
|
| 285 |
+
"learning_rate": 8.020000000000001e-06,
|
| 286 |
+
"loss": 0.4338,
|
| 287 |
+
"step": 1800
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 0.62,
|
| 291 |
+
"learning_rate": 7.686666666666667e-06,
|
| 292 |
+
"loss": 0.5849,
|
| 293 |
+
"step": 1850
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 0.63,
|
| 297 |
+
"learning_rate": 7.353333333333334e-06,
|
| 298 |
+
"loss": 0.6726,
|
| 299 |
+
"step": 1900
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 0.65,
|
| 303 |
+
"learning_rate": 7.0200000000000006e-06,
|
| 304 |
+
"loss": 0.5732,
|
| 305 |
+
"step": 1950
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 0.67,
|
| 309 |
+
"learning_rate": 6.6866666666666665e-06,
|
| 310 |
+
"loss": 0.5251,
|
| 311 |
+
"step": 2000
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.67,
|
| 315 |
+
"eval_accuracy": 0.780448717948718,
|
| 316 |
+
"eval_loss": 0.4892784655094147,
|
| 317 |
+
"eval_runtime": 503.278,
|
| 318 |
+
"eval_samples_per_second": 1.24,
|
| 319 |
+
"eval_steps_per_second": 1.24,
|
| 320 |
+
"step": 2000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.68,
|
| 324 |
+
"learning_rate": 6.353333333333333e-06,
|
| 325 |
+
"loss": 0.6144,
|
| 326 |
+
"step": 2050
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 0.7,
|
| 330 |
+
"learning_rate": 6.02e-06,
|
| 331 |
+
"loss": 0.6247,
|
| 332 |
+
"step": 2100
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.72,
|
| 336 |
+
"learning_rate": 5.6933333333333344e-06,
|
| 337 |
+
"loss": 0.6012,
|
| 338 |
+
"step": 2150
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.73,
|
| 342 |
+
"learning_rate": 5.36e-06,
|
| 343 |
+
"loss": 0.449,
|
| 344 |
+
"step": 2200
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.75,
|
| 348 |
+
"learning_rate": 5.026666666666667e-06,
|
| 349 |
+
"loss": 0.5145,
|
| 350 |
+
"step": 2250
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 0.75,
|
| 354 |
+
"eval_accuracy": 0.7852564102564102,
|
| 355 |
+
"eval_loss": 0.492367684841156,
|
| 356 |
+
"eval_runtime": 502.462,
|
| 357 |
+
"eval_samples_per_second": 1.242,
|
| 358 |
+
"eval_steps_per_second": 1.242,
|
| 359 |
+
"step": 2250
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.77,
|
| 363 |
+
"learning_rate": 4.693333333333334e-06,
|
| 364 |
+
"loss": 0.4815,
|
| 365 |
+
"step": 2300
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 0.78,
|
| 369 |
+
"learning_rate": 4.366666666666667e-06,
|
| 370 |
+
"loss": 0.5096,
|
| 371 |
+
"step": 2350
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 0.8,
|
| 375 |
+
"learning_rate": 4.033333333333333e-06,
|
| 376 |
+
"loss": 0.6567,
|
| 377 |
+
"step": 2400
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.82,
|
| 381 |
+
"learning_rate": 3.7e-06,
|
| 382 |
+
"loss": 0.4821,
|
| 383 |
+
"step": 2450
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 0.83,
|
| 387 |
+
"learning_rate": 3.366666666666667e-06,
|
| 388 |
+
"loss": 0.5085,
|
| 389 |
+
"step": 2500
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 0.83,
|
| 393 |
+
"eval_accuracy": 0.7852564102564102,
|
| 394 |
+
"eval_loss": 0.49340590834617615,
|
| 395 |
+
"eval_runtime": 502.7893,
|
| 396 |
+
"eval_samples_per_second": 1.241,
|
| 397 |
+
"eval_steps_per_second": 1.241,
|
| 398 |
+
"step": 2500
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 0.85,
|
| 402 |
+
"learning_rate": 3.0333333333333337e-06,
|
| 403 |
+
"loss": 0.5839,
|
| 404 |
+
"step": 2550
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 0.87,
|
| 408 |
+
"learning_rate": 2.7000000000000004e-06,
|
| 409 |
+
"loss": 0.6301,
|
| 410 |
+
"step": 2600
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.88,
|
| 414 |
+
"learning_rate": 2.3666666666666667e-06,
|
| 415 |
+
"loss": 0.5963,
|
| 416 |
+
"step": 2650
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.9,
|
| 420 |
+
"learning_rate": 2.0333333333333335e-06,
|
| 421 |
+
"loss": 0.5734,
|
| 422 |
+
"step": 2700
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.92,
|
| 426 |
+
"learning_rate": 1.7000000000000002e-06,
|
| 427 |
+
"loss": 0.617,
|
| 428 |
+
"step": 2750
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 0.92,
|
| 432 |
+
"eval_accuracy": 0.782051282051282,
|
| 433 |
+
"eval_loss": 0.48034408688545227,
|
| 434 |
+
"eval_runtime": 502.6996,
|
| 435 |
+
"eval_samples_per_second": 1.241,
|
| 436 |
+
"eval_steps_per_second": 1.241,
|
| 437 |
+
"step": 2750
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.93,
|
| 441 |
+
"learning_rate": 1.3666666666666668e-06,
|
| 442 |
+
"loss": 0.624,
|
| 443 |
+
"step": 2800
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.95,
|
| 447 |
+
"learning_rate": 1.0333333333333333e-06,
|
| 448 |
+
"loss": 0.5594,
|
| 449 |
+
"step": 2850
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 0.97,
|
| 453 |
+
"learning_rate": 7.000000000000001e-07,
|
| 454 |
+
"loss": 0.527,
|
| 455 |
+
"step": 2900
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 0.98,
|
| 459 |
+
"learning_rate": 3.666666666666667e-07,
|
| 460 |
+
"loss": 0.4313,
|
| 461 |
+
"step": 2950
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 1.0,
|
| 465 |
+
"learning_rate": 3.333333333333334e-08,
|
| 466 |
+
"loss": 0.5525,
|
| 467 |
+
"step": 3000
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 1.0,
|
| 471 |
+
"eval_accuracy": 0.7868589743589743,
|
| 472 |
+
"eval_loss": 0.48102569580078125,
|
| 473 |
+
"eval_runtime": 502.7831,
|
| 474 |
+
"eval_samples_per_second": 1.241,
|
| 475 |
+
"eval_steps_per_second": 1.241,
|
| 476 |
+
"step": 3000
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 1.0,
|
| 480 |
+
"step": 3000,
|
| 481 |
+
"total_flos": 0.0,
|
| 482 |
+
"train_loss": 0.5698420314788818,
|
| 483 |
+
"train_runtime": 30788.306,
|
| 484 |
+
"train_samples_per_second": 0.39,
|
| 485 |
+
"train_steps_per_second": 0.097
|
| 486 |
+
}
|
| 487 |
+
],
|
| 488 |
+
"max_steps": 3000,
|
| 489 |
+
"num_train_epochs": 1,
|
| 490 |
+
"total_flos": 0.0,
|
| 491 |
+
"trial_name": null,
|
| 492 |
+
"trial_params": null
|
| 493 |
+
}
|