Commit
·
2ce838e
1
Parent(s):
833888d
End of training
Browse files- README.md +13 -0
- trainer_state.json +502 -0
README.md
CHANGED
|
@@ -5,6 +5,7 @@ tags:
|
|
| 5 |
model-index:
|
| 6 |
- name: llama-2-7b-reward-oasst1
|
| 7 |
results: []
|
|
|
|
| 8 |
---
|
| 9 |
|
| 10 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -36,6 +37,17 @@ More information needed
|
|
| 36 |
|
| 37 |
## Training procedure
|
| 38 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 39 |
### Training hyperparameters
|
| 40 |
|
| 41 |
The following hyperparameters were used during training:
|
|
@@ -51,6 +63,7 @@ The following hyperparameters were used during training:
|
|
| 51 |
|
| 52 |
### Framework versions
|
| 53 |
|
|
|
|
| 54 |
- Transformers 4.32.0.dev0
|
| 55 |
- Pytorch 2.0.1+cu118
|
| 56 |
- Datasets 2.14.0
|
|
|
|
| 5 |
model-index:
|
| 6 |
- name: llama-2-7b-reward-oasst1
|
| 7 |
results: []
|
| 8 |
+
library_name: peft
|
| 9 |
---
|
| 10 |
|
| 11 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 37 |
|
| 38 |
## Training procedure
|
| 39 |
|
| 40 |
+
|
| 41 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 42 |
+
- load_in_8bit: False
|
| 43 |
+
- load_in_4bit: True
|
| 44 |
+
- llm_int8_threshold: 6.0
|
| 45 |
+
- llm_int8_skip_modules: None
|
| 46 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 47 |
+
- llm_int8_has_fp16_weight: False
|
| 48 |
+
- bnb_4bit_quant_type: nf4
|
| 49 |
+
- bnb_4bit_use_double_quant: False
|
| 50 |
+
- bnb_4bit_compute_dtype: float16
|
| 51 |
### Training hyperparameters
|
| 52 |
|
| 53 |
The following hyperparameters were used during training:
|
|
|
|
| 63 |
|
| 64 |
### Framework versions
|
| 65 |
|
| 66 |
+
- PEFT 0.5.0.dev0
|
| 67 |
- Transformers 4.32.0.dev0
|
| 68 |
- Pytorch 2.0.1+cu118
|
| 69 |
- Datasets 2.14.0
|
trainer_state.json
ADDED
|
@@ -0,0 +1,502 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.7561820571018995,
|
| 5 |
+
"global_step": 3165,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.01,
|
| 12 |
+
"learning_rate": 1.978494623655914e-05,
|
| 13 |
+
"loss": 1.0308,
|
| 14 |
+
"step": 50
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.02,
|
| 18 |
+
"learning_rate": 1.954599761051374e-05,
|
| 19 |
+
"loss": 0.7656,
|
| 20 |
+
"step": 100
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.04,
|
| 24 |
+
"learning_rate": 1.930704898446834e-05,
|
| 25 |
+
"loss": 0.8032,
|
| 26 |
+
"step": 150
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.05,
|
| 30 |
+
"learning_rate": 1.906810035842294e-05,
|
| 31 |
+
"loss": 0.7398,
|
| 32 |
+
"step": 200
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.06,
|
| 36 |
+
"learning_rate": 1.882915173237754e-05,
|
| 37 |
+
"loss": 0.7004,
|
| 38 |
+
"step": 250
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.06,
|
| 42 |
+
"eval_accuracy": 0.6535433070866141,
|
| 43 |
+
"eval_loss": 0.7174085378646851,
|
| 44 |
+
"eval_runtime": 431.9798,
|
| 45 |
+
"eval_samples_per_second": 2.058,
|
| 46 |
+
"eval_steps_per_second": 2.058,
|
| 47 |
+
"step": 250
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 0.07,
|
| 51 |
+
"learning_rate": 1.859020310633214e-05,
|
| 52 |
+
"loss": 0.8009,
|
| 53 |
+
"step": 300
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.08,
|
| 57 |
+
"learning_rate": 1.835125448028674e-05,
|
| 58 |
+
"loss": 0.6551,
|
| 59 |
+
"step": 350
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.1,
|
| 63 |
+
"learning_rate": 1.811230585424134e-05,
|
| 64 |
+
"loss": 0.6348,
|
| 65 |
+
"step": 400
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.11,
|
| 69 |
+
"learning_rate": 1.787335722819594e-05,
|
| 70 |
+
"loss": 0.5236,
|
| 71 |
+
"step": 450
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 0.12,
|
| 75 |
+
"learning_rate": 1.763440860215054e-05,
|
| 76 |
+
"loss": 0.6735,
|
| 77 |
+
"step": 500
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 0.12,
|
| 81 |
+
"eval_accuracy": 0.7007874015748031,
|
| 82 |
+
"eval_loss": 0.6178467869758606,
|
| 83 |
+
"eval_runtime": 430.5762,
|
| 84 |
+
"eval_samples_per_second": 2.065,
|
| 85 |
+
"eval_steps_per_second": 2.065,
|
| 86 |
+
"step": 500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.13,
|
| 90 |
+
"learning_rate": 1.7395459976105136e-05,
|
| 91 |
+
"loss": 0.7371,
|
| 92 |
+
"step": 550
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.14,
|
| 96 |
+
"learning_rate": 1.7156511350059738e-05,
|
| 97 |
+
"loss": 0.6957,
|
| 98 |
+
"step": 600
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.16,
|
| 102 |
+
"learning_rate": 1.691756272401434e-05,
|
| 103 |
+
"loss": 0.592,
|
| 104 |
+
"step": 650
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.17,
|
| 108 |
+
"learning_rate": 1.6678614097968937e-05,
|
| 109 |
+
"loss": 0.5799,
|
| 110 |
+
"step": 700
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.18,
|
| 114 |
+
"learning_rate": 1.6439665471923538e-05,
|
| 115 |
+
"loss": 0.6165,
|
| 116 |
+
"step": 750
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.18,
|
| 120 |
+
"eval_accuracy": 0.7424071991001124,
|
| 121 |
+
"eval_loss": 0.5588846206665039,
|
| 122 |
+
"eval_runtime": 431.475,
|
| 123 |
+
"eval_samples_per_second": 2.06,
|
| 124 |
+
"eval_steps_per_second": 2.06,
|
| 125 |
+
"step": 750
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 0.19,
|
| 129 |
+
"learning_rate": 1.6200716845878136e-05,
|
| 130 |
+
"loss": 0.5045,
|
| 131 |
+
"step": 800
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 0.2,
|
| 135 |
+
"learning_rate": 1.5961768219832737e-05,
|
| 136 |
+
"loss": 0.6213,
|
| 137 |
+
"step": 850
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.22,
|
| 141 |
+
"learning_rate": 1.5722819593787338e-05,
|
| 142 |
+
"loss": 0.6248,
|
| 143 |
+
"step": 900
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.23,
|
| 147 |
+
"learning_rate": 1.5483870967741936e-05,
|
| 148 |
+
"loss": 0.4998,
|
| 149 |
+
"step": 950
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.24,
|
| 153 |
+
"learning_rate": 1.5244922341696537e-05,
|
| 154 |
+
"loss": 0.6603,
|
| 155 |
+
"step": 1000
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 0.24,
|
| 159 |
+
"eval_accuracy": 0.7559055118110236,
|
| 160 |
+
"eval_loss": 0.5710476636886597,
|
| 161 |
+
"eval_runtime": 432.0116,
|
| 162 |
+
"eval_samples_per_second": 2.058,
|
| 163 |
+
"eval_steps_per_second": 2.058,
|
| 164 |
+
"step": 1000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.25,
|
| 168 |
+
"learning_rate": 1.5005973715651137e-05,
|
| 169 |
+
"loss": 0.5468,
|
| 170 |
+
"step": 1050
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.26,
|
| 174 |
+
"learning_rate": 1.4767025089605736e-05,
|
| 175 |
+
"loss": 0.511,
|
| 176 |
+
"step": 1100
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.27,
|
| 180 |
+
"learning_rate": 1.4528076463560337e-05,
|
| 181 |
+
"loss": 0.5575,
|
| 182 |
+
"step": 1150
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.29,
|
| 186 |
+
"learning_rate": 1.4289127837514935e-05,
|
| 187 |
+
"loss": 0.7658,
|
| 188 |
+
"step": 1200
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.3,
|
| 192 |
+
"learning_rate": 1.4050179211469535e-05,
|
| 193 |
+
"loss": 0.5575,
|
| 194 |
+
"step": 1250
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.3,
|
| 198 |
+
"eval_accuracy": 0.7570303712035995,
|
| 199 |
+
"eval_loss": 0.5421488881111145,
|
| 200 |
+
"eval_runtime": 430.448,
|
| 201 |
+
"eval_samples_per_second": 2.065,
|
| 202 |
+
"eval_steps_per_second": 2.065,
|
| 203 |
+
"step": 1250
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 0.31,
|
| 207 |
+
"learning_rate": 1.3811230585424136e-05,
|
| 208 |
+
"loss": 0.6252,
|
| 209 |
+
"step": 1300
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.32,
|
| 213 |
+
"learning_rate": 1.3572281959378735e-05,
|
| 214 |
+
"loss": 0.4673,
|
| 215 |
+
"step": 1350
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 0.33,
|
| 219 |
+
"learning_rate": 1.3333333333333333e-05,
|
| 220 |
+
"loss": 0.5266,
|
| 221 |
+
"step": 1400
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 0.35,
|
| 225 |
+
"learning_rate": 1.3094384707287935e-05,
|
| 226 |
+
"loss": 0.6353,
|
| 227 |
+
"step": 1450
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.36,
|
| 231 |
+
"learning_rate": 1.2855436081242534e-05,
|
| 232 |
+
"loss": 0.589,
|
| 233 |
+
"step": 1500
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.36,
|
| 237 |
+
"eval_accuracy": 0.750281214848144,
|
| 238 |
+
"eval_loss": 0.5329739451408386,
|
| 239 |
+
"eval_runtime": 429.8372,
|
| 240 |
+
"eval_samples_per_second": 2.068,
|
| 241 |
+
"eval_steps_per_second": 2.068,
|
| 242 |
+
"step": 1500
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.37,
|
| 246 |
+
"learning_rate": 1.2616487455197134e-05,
|
| 247 |
+
"loss": 0.6944,
|
| 248 |
+
"step": 1550
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.38,
|
| 252 |
+
"learning_rate": 1.2377538829151735e-05,
|
| 253 |
+
"loss": 0.4992,
|
| 254 |
+
"step": 1600
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.39,
|
| 258 |
+
"learning_rate": 1.2138590203106333e-05,
|
| 259 |
+
"loss": 0.6429,
|
| 260 |
+
"step": 1650
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.41,
|
| 264 |
+
"learning_rate": 1.1899641577060932e-05,
|
| 265 |
+
"loss": 0.5243,
|
| 266 |
+
"step": 1700
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 0.42,
|
| 270 |
+
"learning_rate": 1.1660692951015533e-05,
|
| 271 |
+
"loss": 0.5644,
|
| 272 |
+
"step": 1750
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.42,
|
| 276 |
+
"eval_accuracy": 0.7559055118110236,
|
| 277 |
+
"eval_loss": 0.5390347838401794,
|
| 278 |
+
"eval_runtime": 429.8773,
|
| 279 |
+
"eval_samples_per_second": 2.068,
|
| 280 |
+
"eval_steps_per_second": 2.068,
|
| 281 |
+
"step": 1750
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 0.43,
|
| 285 |
+
"learning_rate": 1.1421744324970133e-05,
|
| 286 |
+
"loss": 0.6321,
|
| 287 |
+
"step": 1800
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 0.44,
|
| 291 |
+
"learning_rate": 1.118279569892473e-05,
|
| 292 |
+
"loss": 0.5939,
|
| 293 |
+
"step": 1850
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 0.45,
|
| 297 |
+
"learning_rate": 1.0943847072879332e-05,
|
| 298 |
+
"loss": 0.5477,
|
| 299 |
+
"step": 1900
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 0.47,
|
| 303 |
+
"learning_rate": 1.0704898446833931e-05,
|
| 304 |
+
"loss": 0.5171,
|
| 305 |
+
"step": 1950
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 0.48,
|
| 309 |
+
"learning_rate": 1.0465949820788533e-05,
|
| 310 |
+
"loss": 0.503,
|
| 311 |
+
"step": 2000
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.48,
|
| 315 |
+
"eval_accuracy": 0.7592800899887514,
|
| 316 |
+
"eval_loss": 0.5502843856811523,
|
| 317 |
+
"eval_runtime": 429.7289,
|
| 318 |
+
"eval_samples_per_second": 2.069,
|
| 319 |
+
"eval_steps_per_second": 2.069,
|
| 320 |
+
"step": 2000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.49,
|
| 324 |
+
"learning_rate": 1.0227001194743132e-05,
|
| 325 |
+
"loss": 0.6059,
|
| 326 |
+
"step": 2050
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 0.5,
|
| 330 |
+
"learning_rate": 9.98805256869773e-06,
|
| 331 |
+
"loss": 0.5716,
|
| 332 |
+
"step": 2100
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.51,
|
| 336 |
+
"learning_rate": 9.749103942652331e-06,
|
| 337 |
+
"loss": 0.5367,
|
| 338 |
+
"step": 2150
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.53,
|
| 342 |
+
"learning_rate": 9.51015531660693e-06,
|
| 343 |
+
"loss": 0.578,
|
| 344 |
+
"step": 2200
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.54,
|
| 348 |
+
"learning_rate": 9.27120669056153e-06,
|
| 349 |
+
"loss": 0.6361,
|
| 350 |
+
"step": 2250
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 0.54,
|
| 354 |
+
"eval_accuracy": 0.7637795275590551,
|
| 355 |
+
"eval_loss": 0.5347180366516113,
|
| 356 |
+
"eval_runtime": 432.0598,
|
| 357 |
+
"eval_samples_per_second": 2.058,
|
| 358 |
+
"eval_steps_per_second": 2.058,
|
| 359 |
+
"step": 2250
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.55,
|
| 363 |
+
"learning_rate": 9.03225806451613e-06,
|
| 364 |
+
"loss": 0.5387,
|
| 365 |
+
"step": 2300
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 0.56,
|
| 369 |
+
"learning_rate": 8.793309438470729e-06,
|
| 370 |
+
"loss": 0.5409,
|
| 371 |
+
"step": 2350
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 0.57,
|
| 375 |
+
"learning_rate": 8.55436081242533e-06,
|
| 376 |
+
"loss": 0.4895,
|
| 377 |
+
"step": 2400
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.59,
|
| 381 |
+
"learning_rate": 8.315412186379928e-06,
|
| 382 |
+
"loss": 0.5966,
|
| 383 |
+
"step": 2450
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 0.6,
|
| 387 |
+
"learning_rate": 8.07646356033453e-06,
|
| 388 |
+
"loss": 0.517,
|
| 389 |
+
"step": 2500
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 0.6,
|
| 393 |
+
"eval_accuracy": 0.7649043869516311,
|
| 394 |
+
"eval_loss": 0.5409161448478699,
|
| 395 |
+
"eval_runtime": 432.0753,
|
| 396 |
+
"eval_samples_per_second": 2.058,
|
| 397 |
+
"eval_steps_per_second": 2.058,
|
| 398 |
+
"step": 2500
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 0.61,
|
| 402 |
+
"learning_rate": 7.837514934289129e-06,
|
| 403 |
+
"loss": 0.5852,
|
| 404 |
+
"step": 2550
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 0.62,
|
| 408 |
+
"learning_rate": 7.5985663082437275e-06,
|
| 409 |
+
"loss": 0.5853,
|
| 410 |
+
"step": 2600
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.63,
|
| 414 |
+
"learning_rate": 7.359617682198328e-06,
|
| 415 |
+
"loss": 0.6096,
|
| 416 |
+
"step": 2650
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.65,
|
| 420 |
+
"learning_rate": 7.120669056152928e-06,
|
| 421 |
+
"loss": 0.6285,
|
| 422 |
+
"step": 2700
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.66,
|
| 426 |
+
"learning_rate": 6.881720430107528e-06,
|
| 427 |
+
"loss": 0.5481,
|
| 428 |
+
"step": 2750
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 0.66,
|
| 432 |
+
"eval_accuracy": 0.7525309336332958,
|
| 433 |
+
"eval_loss": 0.5184136033058167,
|
| 434 |
+
"eval_runtime": 431.9074,
|
| 435 |
+
"eval_samples_per_second": 2.058,
|
| 436 |
+
"eval_steps_per_second": 2.058,
|
| 437 |
+
"step": 2750
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.67,
|
| 441 |
+
"learning_rate": 6.642771804062127e-06,
|
| 442 |
+
"loss": 0.6294,
|
| 443 |
+
"step": 2800
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.68,
|
| 447 |
+
"learning_rate": 6.403823178016727e-06,
|
| 448 |
+
"loss": 0.5585,
|
| 449 |
+
"step": 2850
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 0.69,
|
| 453 |
+
"learning_rate": 6.164874551971327e-06,
|
| 454 |
+
"loss": 0.5914,
|
| 455 |
+
"step": 2900
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 0.7,
|
| 459 |
+
"learning_rate": 5.925925925925926e-06,
|
| 460 |
+
"loss": 0.5831,
|
| 461 |
+
"step": 2950
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 0.72,
|
| 465 |
+
"learning_rate": 5.686977299880526e-06,
|
| 466 |
+
"loss": 0.6036,
|
| 467 |
+
"step": 3000
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 0.72,
|
| 471 |
+
"eval_accuracy": 0.7592800899887514,
|
| 472 |
+
"eval_loss": 0.5061925649642944,
|
| 473 |
+
"eval_runtime": 431.9996,
|
| 474 |
+
"eval_samples_per_second": 2.058,
|
| 475 |
+
"eval_steps_per_second": 2.058,
|
| 476 |
+
"step": 3000
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 0.73,
|
| 480 |
+
"learning_rate": 5.4480286738351265e-06,
|
| 481 |
+
"loss": 0.4684,
|
| 482 |
+
"step": 3050
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 0.74,
|
| 486 |
+
"learning_rate": 5.209080047789725e-06,
|
| 487 |
+
"loss": 0.5459,
|
| 488 |
+
"step": 3100
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 0.75,
|
| 492 |
+
"learning_rate": 4.9701314217443256e-06,
|
| 493 |
+
"loss": 0.5358,
|
| 494 |
+
"step": 3150
|
| 495 |
+
}
|
| 496 |
+
],
|
| 497 |
+
"max_steps": 4185,
|
| 498 |
+
"num_train_epochs": 1,
|
| 499 |
+
"total_flos": 0.0,
|
| 500 |
+
"trial_name": null,
|
| 501 |
+
"trial_params": null
|
| 502 |
+
}
|