Upload folder using huggingface_hub
Browse files- checkpoint-108/adapter_config.json +19 -0
- checkpoint-108/adapter_model.bin +3 -0
- checkpoint-108/optimizer.pt +3 -0
- checkpoint-108/rng_state.pth +3 -0
- checkpoint-108/scheduler.pt +3 -0
- checkpoint-108/trainer_state.json +297 -0
- checkpoint-108/training_args.bin +3 -0
- checkpoint-120/adapter_config.json +19 -0
- checkpoint-120/adapter_model.bin +3 -0
- checkpoint-120/optimizer.pt +3 -0
- checkpoint-120/rng_state.pth +3 -0
- checkpoint-120/scheduler.pt +3 -0
- checkpoint-120/trainer_state.json +333 -0
- checkpoint-120/training_args.bin +3 -0
- checkpoint-96/adapter_config.json +19 -0
- checkpoint-96/adapter_model.bin +3 -0
- checkpoint-96/optimizer.pt +3 -0
- checkpoint-96/rng_state.pth +3 -0
- checkpoint-96/scheduler.pt +3 -0
- checkpoint-96/trainer_state.json +267 -0
- checkpoint-96/training_args.bin +3 -0
checkpoint-108/adapter_config.json
ADDED
|
@@ -0,0 +1,19 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"base_model_name_or_path": "codellama/CodeLlama-7b-hf",
|
| 3 |
+
"bias": "none",
|
| 4 |
+
"enable_lora": null,
|
| 5 |
+
"fan_in_fan_out": false,
|
| 6 |
+
"inference_mode": true,
|
| 7 |
+
"init_lora_weights": true,
|
| 8 |
+
"lora_alpha": 16,
|
| 9 |
+
"lora_dropout": 0.05,
|
| 10 |
+
"merge_weights": false,
|
| 11 |
+
"modules_to_save": null,
|
| 12 |
+
"peft_type": "LORA",
|
| 13 |
+
"r": 8,
|
| 14 |
+
"target_modules": [
|
| 15 |
+
"q_proj",
|
| 16 |
+
"v_proj"
|
| 17 |
+
],
|
| 18 |
+
"task_type": "CAUSAL_LM"
|
| 19 |
+
}
|
checkpoint-108/adapter_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2f9050ee62fd25a9062f57f81a90e27e01c501d7ccc6f10786f762e66fe32520
|
| 3 |
+
size 16822989
|
checkpoint-108/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d5e51cdc64e56f277b8e62ef82b5c783f1ac01825ad8d9e8906eb52bcfb0f3b1
|
| 3 |
+
size 33661637
|
checkpoint-108/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c2d31d965312af96ad728bd0641d6d3f8b83ba520beadb760402d53868041d7d
|
| 3 |
+
size 14575
|
checkpoint-108/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d2050e9766ba8af43a09060f4537765109e5253e38bb78cac58004e495ea42c4
|
| 3 |
+
size 627
|
checkpoint-108/trainer_state.json
ADDED
|
@@ -0,0 +1,297 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.14893724024295807,
|
| 3 |
+
"best_model_checkpoint": "./lora-out/checkpoint-108",
|
| 4 |
+
"epoch": 7.646017699115045,
|
| 5 |
+
"eval_steps": 4,
|
| 6 |
+
"global_step": 108,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.28,
|
| 13 |
+
"eval_loss": 0.3006412982940674,
|
| 14 |
+
"eval_runtime": 54.7333,
|
| 15 |
+
"eval_samples_per_second": 0.914,
|
| 16 |
+
"eval_steps_per_second": 0.238,
|
| 17 |
+
"step": 4
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.57,
|
| 21 |
+
"eval_loss": 0.300335168838501,
|
| 22 |
+
"eval_runtime": 54.8113,
|
| 23 |
+
"eval_samples_per_second": 0.912,
|
| 24 |
+
"eval_steps_per_second": 0.237,
|
| 25 |
+
"step": 8
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.71,
|
| 29 |
+
"learning_rate": 1e-05,
|
| 30 |
+
"loss": 0.3024,
|
| 31 |
+
"step": 10
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.85,
|
| 35 |
+
"eval_loss": 0.2993900179862976,
|
| 36 |
+
"eval_runtime": 54.782,
|
| 37 |
+
"eval_samples_per_second": 0.913,
|
| 38 |
+
"eval_steps_per_second": 0.237,
|
| 39 |
+
"step": 12
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 1.13,
|
| 43 |
+
"eval_loss": 0.29816487431526184,
|
| 44 |
+
"eval_runtime": 54.8049,
|
| 45 |
+
"eval_samples_per_second": 0.912,
|
| 46 |
+
"eval_steps_per_second": 0.237,
|
| 47 |
+
"step": 16
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 1.42,
|
| 51 |
+
"learning_rate": 2e-05,
|
| 52 |
+
"loss": 0.3035,
|
| 53 |
+
"step": 20
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.42,
|
| 57 |
+
"eval_loss": 0.29595255851745605,
|
| 58 |
+
"eval_runtime": 54.7885,
|
| 59 |
+
"eval_samples_per_second": 0.913,
|
| 60 |
+
"eval_steps_per_second": 0.237,
|
| 61 |
+
"step": 20
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 1.7,
|
| 65 |
+
"eval_loss": 0.2939557135105133,
|
| 66 |
+
"eval_runtime": 54.7999,
|
| 67 |
+
"eval_samples_per_second": 0.912,
|
| 68 |
+
"eval_steps_per_second": 0.237,
|
| 69 |
+
"step": 24
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 1.98,
|
| 73 |
+
"eval_loss": 0.29013773798942566,
|
| 74 |
+
"eval_runtime": 54.7805,
|
| 75 |
+
"eval_samples_per_second": 0.913,
|
| 76 |
+
"eval_steps_per_second": 0.237,
|
| 77 |
+
"step": 28
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 2.12,
|
| 81 |
+
"learning_rate": 3e-05,
|
| 82 |
+
"loss": 0.2959,
|
| 83 |
+
"step": 30
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 2.27,
|
| 87 |
+
"eval_loss": 0.28251081705093384,
|
| 88 |
+
"eval_runtime": 54.7706,
|
| 89 |
+
"eval_samples_per_second": 0.913,
|
| 90 |
+
"eval_steps_per_second": 0.237,
|
| 91 |
+
"step": 32
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 2.55,
|
| 95 |
+
"eval_loss": 0.2771329879760742,
|
| 96 |
+
"eval_runtime": 54.7818,
|
| 97 |
+
"eval_samples_per_second": 0.913,
|
| 98 |
+
"eval_steps_per_second": 0.237,
|
| 99 |
+
"step": 36
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 2.83,
|
| 103 |
+
"learning_rate": 4e-05,
|
| 104 |
+
"loss": 0.284,
|
| 105 |
+
"step": 40
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 2.83,
|
| 109 |
+
"eval_loss": 0.27146488428115845,
|
| 110 |
+
"eval_runtime": 54.803,
|
| 111 |
+
"eval_samples_per_second": 0.912,
|
| 112 |
+
"eval_steps_per_second": 0.237,
|
| 113 |
+
"step": 40
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 3.12,
|
| 117 |
+
"eval_loss": 0.26464152336120605,
|
| 118 |
+
"eval_runtime": 54.8467,
|
| 119 |
+
"eval_samples_per_second": 0.912,
|
| 120 |
+
"eval_steps_per_second": 0.237,
|
| 121 |
+
"step": 44
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 3.4,
|
| 125 |
+
"eval_loss": 0.25653430819511414,
|
| 126 |
+
"eval_runtime": 54.8327,
|
| 127 |
+
"eval_samples_per_second": 0.912,
|
| 128 |
+
"eval_steps_per_second": 0.237,
|
| 129 |
+
"step": 48
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 3.54,
|
| 133 |
+
"learning_rate": 5e-05,
|
| 134 |
+
"loss": 0.263,
|
| 135 |
+
"step": 50
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 3.68,
|
| 139 |
+
"eval_loss": 0.24627122282981873,
|
| 140 |
+
"eval_runtime": 54.813,
|
| 141 |
+
"eval_samples_per_second": 0.912,
|
| 142 |
+
"eval_steps_per_second": 0.237,
|
| 143 |
+
"step": 52
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 3.96,
|
| 147 |
+
"eval_loss": 0.23474617302417755,
|
| 148 |
+
"eval_runtime": 54.7901,
|
| 149 |
+
"eval_samples_per_second": 0.913,
|
| 150 |
+
"eval_steps_per_second": 0.237,
|
| 151 |
+
"step": 56
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 4.25,
|
| 155 |
+
"learning_rate": 6e-05,
|
| 156 |
+
"loss": 0.241,
|
| 157 |
+
"step": 60
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 4.25,
|
| 161 |
+
"eval_loss": 0.2220366895198822,
|
| 162 |
+
"eval_runtime": 54.7983,
|
| 163 |
+
"eval_samples_per_second": 0.912,
|
| 164 |
+
"eval_steps_per_second": 0.237,
|
| 165 |
+
"step": 60
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 4.53,
|
| 169 |
+
"eval_loss": 0.20926769077777863,
|
| 170 |
+
"eval_runtime": 54.7403,
|
| 171 |
+
"eval_samples_per_second": 0.913,
|
| 172 |
+
"eval_steps_per_second": 0.237,
|
| 173 |
+
"step": 64
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 4.81,
|
| 177 |
+
"eval_loss": 0.19629451632499695,
|
| 178 |
+
"eval_runtime": 54.7525,
|
| 179 |
+
"eval_samples_per_second": 0.913,
|
| 180 |
+
"eval_steps_per_second": 0.237,
|
| 181 |
+
"step": 68
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 4.96,
|
| 185 |
+
"learning_rate": 7e-05,
|
| 186 |
+
"loss": 0.2101,
|
| 187 |
+
"step": 70
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 5.1,
|
| 191 |
+
"eval_loss": 0.18524658679962158,
|
| 192 |
+
"eval_runtime": 54.7303,
|
| 193 |
+
"eval_samples_per_second": 0.914,
|
| 194 |
+
"eval_steps_per_second": 0.238,
|
| 195 |
+
"step": 72
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 5.38,
|
| 199 |
+
"eval_loss": 0.17731742560863495,
|
| 200 |
+
"eval_runtime": 54.7552,
|
| 201 |
+
"eval_samples_per_second": 0.913,
|
| 202 |
+
"eval_steps_per_second": 0.237,
|
| 203 |
+
"step": 76
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 5.66,
|
| 207 |
+
"learning_rate": 8e-05,
|
| 208 |
+
"loss": 0.1788,
|
| 209 |
+
"step": 80
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 5.66,
|
| 213 |
+
"eval_loss": 0.16993452608585358,
|
| 214 |
+
"eval_runtime": 54.729,
|
| 215 |
+
"eval_samples_per_second": 0.914,
|
| 216 |
+
"eval_steps_per_second": 0.238,
|
| 217 |
+
"step": 80
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 5.95,
|
| 221 |
+
"eval_loss": 0.164781853556633,
|
| 222 |
+
"eval_runtime": 54.741,
|
| 223 |
+
"eval_samples_per_second": 0.913,
|
| 224 |
+
"eval_steps_per_second": 0.237,
|
| 225 |
+
"step": 84
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 6.23,
|
| 229 |
+
"eval_loss": 0.16103117167949677,
|
| 230 |
+
"eval_runtime": 54.7837,
|
| 231 |
+
"eval_samples_per_second": 0.913,
|
| 232 |
+
"eval_steps_per_second": 0.237,
|
| 233 |
+
"step": 88
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 6.37,
|
| 237 |
+
"learning_rate": 9e-05,
|
| 238 |
+
"loss": 0.1615,
|
| 239 |
+
"step": 90
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 6.51,
|
| 243 |
+
"eval_loss": 0.15781742334365845,
|
| 244 |
+
"eval_runtime": 54.7138,
|
| 245 |
+
"eval_samples_per_second": 0.914,
|
| 246 |
+
"eval_steps_per_second": 0.238,
|
| 247 |
+
"step": 92
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 6.8,
|
| 251 |
+
"eval_loss": 0.15516981482505798,
|
| 252 |
+
"eval_runtime": 54.7516,
|
| 253 |
+
"eval_samples_per_second": 0.913,
|
| 254 |
+
"eval_steps_per_second": 0.237,
|
| 255 |
+
"step": 96
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 7.08,
|
| 259 |
+
"learning_rate": 0.0001,
|
| 260 |
+
"loss": 0.1533,
|
| 261 |
+
"step": 100
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 7.08,
|
| 265 |
+
"eval_loss": 0.15261690318584442,
|
| 266 |
+
"eval_runtime": 54.6891,
|
| 267 |
+
"eval_samples_per_second": 0.914,
|
| 268 |
+
"eval_steps_per_second": 0.238,
|
| 269 |
+
"step": 100
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 7.36,
|
| 273 |
+
"eval_loss": 0.15066812932491302,
|
| 274 |
+
"eval_runtime": 54.6884,
|
| 275 |
+
"eval_samples_per_second": 0.914,
|
| 276 |
+
"eval_steps_per_second": 0.238,
|
| 277 |
+
"step": 104
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 7.65,
|
| 281 |
+
"eval_loss": 0.14893724024295807,
|
| 282 |
+
"eval_runtime": 54.6275,
|
| 283 |
+
"eval_samples_per_second": 0.915,
|
| 284 |
+
"eval_steps_per_second": 0.238,
|
| 285 |
+
"step": 108
|
| 286 |
+
}
|
| 287 |
+
],
|
| 288 |
+
"logging_steps": 10,
|
| 289 |
+
"max_steps": 210,
|
| 290 |
+
"num_input_tokens_seen": 0,
|
| 291 |
+
"num_train_epochs": 15,
|
| 292 |
+
"save_steps": 12,
|
| 293 |
+
"total_flos": 3.553655402645422e+17,
|
| 294 |
+
"train_batch_size": 4,
|
| 295 |
+
"trial_name": null,
|
| 296 |
+
"trial_params": null
|
| 297 |
+
}
|
checkpoint-108/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:85310c54a4f279d40e8badbc8f6f7406b57e15fc0c79500de525827feedf5072
|
| 3 |
+
size 4219
|
checkpoint-120/adapter_config.json
ADDED
|
@@ -0,0 +1,19 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"base_model_name_or_path": "codellama/CodeLlama-7b-hf",
|
| 3 |
+
"bias": "none",
|
| 4 |
+
"enable_lora": null,
|
| 5 |
+
"fan_in_fan_out": false,
|
| 6 |
+
"inference_mode": true,
|
| 7 |
+
"init_lora_weights": true,
|
| 8 |
+
"lora_alpha": 16,
|
| 9 |
+
"lora_dropout": 0.05,
|
| 10 |
+
"merge_weights": false,
|
| 11 |
+
"modules_to_save": null,
|
| 12 |
+
"peft_type": "LORA",
|
| 13 |
+
"r": 8,
|
| 14 |
+
"target_modules": [
|
| 15 |
+
"q_proj",
|
| 16 |
+
"v_proj"
|
| 17 |
+
],
|
| 18 |
+
"task_type": "CAUSAL_LM"
|
| 19 |
+
}
|
checkpoint-120/adapter_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ab2a1e562884c0574f07ef80b399f6f2536bc226604adc506df3776c8477bde0
|
| 3 |
+
size 16822989
|
checkpoint-120/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c688a73e9177c4c85cb2171a6fcd0043885669c2e8f6aee958e16e92e9e706d7
|
| 3 |
+
size 33661637
|
checkpoint-120/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5eb8ebeef2d7062f8918cac9d79f55abf7a0736d6e439d85b087b02f5b32707d
|
| 3 |
+
size 14575
|
checkpoint-120/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:faeed8541dd712b0b41ff7783005b44f5bb694e062c68740a92308cb1f7e61f5
|
| 3 |
+
size 627
|
checkpoint-120/trainer_state.json
ADDED
|
@@ -0,0 +1,333 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.14450186491012573,
|
| 3 |
+
"best_model_checkpoint": "./lora-out/checkpoint-120",
|
| 4 |
+
"epoch": 8.495575221238939,
|
| 5 |
+
"eval_steps": 4,
|
| 6 |
+
"global_step": 120,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.28,
|
| 13 |
+
"eval_loss": 0.3006412982940674,
|
| 14 |
+
"eval_runtime": 54.7333,
|
| 15 |
+
"eval_samples_per_second": 0.914,
|
| 16 |
+
"eval_steps_per_second": 0.238,
|
| 17 |
+
"step": 4
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.57,
|
| 21 |
+
"eval_loss": 0.300335168838501,
|
| 22 |
+
"eval_runtime": 54.8113,
|
| 23 |
+
"eval_samples_per_second": 0.912,
|
| 24 |
+
"eval_steps_per_second": 0.237,
|
| 25 |
+
"step": 8
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.71,
|
| 29 |
+
"learning_rate": 1e-05,
|
| 30 |
+
"loss": 0.3024,
|
| 31 |
+
"step": 10
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.85,
|
| 35 |
+
"eval_loss": 0.2993900179862976,
|
| 36 |
+
"eval_runtime": 54.782,
|
| 37 |
+
"eval_samples_per_second": 0.913,
|
| 38 |
+
"eval_steps_per_second": 0.237,
|
| 39 |
+
"step": 12
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 1.13,
|
| 43 |
+
"eval_loss": 0.29816487431526184,
|
| 44 |
+
"eval_runtime": 54.8049,
|
| 45 |
+
"eval_samples_per_second": 0.912,
|
| 46 |
+
"eval_steps_per_second": 0.237,
|
| 47 |
+
"step": 16
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 1.42,
|
| 51 |
+
"learning_rate": 2e-05,
|
| 52 |
+
"loss": 0.3035,
|
| 53 |
+
"step": 20
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.42,
|
| 57 |
+
"eval_loss": 0.29595255851745605,
|
| 58 |
+
"eval_runtime": 54.7885,
|
| 59 |
+
"eval_samples_per_second": 0.913,
|
| 60 |
+
"eval_steps_per_second": 0.237,
|
| 61 |
+
"step": 20
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 1.7,
|
| 65 |
+
"eval_loss": 0.2939557135105133,
|
| 66 |
+
"eval_runtime": 54.7999,
|
| 67 |
+
"eval_samples_per_second": 0.912,
|
| 68 |
+
"eval_steps_per_second": 0.237,
|
| 69 |
+
"step": 24
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 1.98,
|
| 73 |
+
"eval_loss": 0.29013773798942566,
|
| 74 |
+
"eval_runtime": 54.7805,
|
| 75 |
+
"eval_samples_per_second": 0.913,
|
| 76 |
+
"eval_steps_per_second": 0.237,
|
| 77 |
+
"step": 28
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 2.12,
|
| 81 |
+
"learning_rate": 3e-05,
|
| 82 |
+
"loss": 0.2959,
|
| 83 |
+
"step": 30
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 2.27,
|
| 87 |
+
"eval_loss": 0.28251081705093384,
|
| 88 |
+
"eval_runtime": 54.7706,
|
| 89 |
+
"eval_samples_per_second": 0.913,
|
| 90 |
+
"eval_steps_per_second": 0.237,
|
| 91 |
+
"step": 32
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 2.55,
|
| 95 |
+
"eval_loss": 0.2771329879760742,
|
| 96 |
+
"eval_runtime": 54.7818,
|
| 97 |
+
"eval_samples_per_second": 0.913,
|
| 98 |
+
"eval_steps_per_second": 0.237,
|
| 99 |
+
"step": 36
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 2.83,
|
| 103 |
+
"learning_rate": 4e-05,
|
| 104 |
+
"loss": 0.284,
|
| 105 |
+
"step": 40
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 2.83,
|
| 109 |
+
"eval_loss": 0.27146488428115845,
|
| 110 |
+
"eval_runtime": 54.803,
|
| 111 |
+
"eval_samples_per_second": 0.912,
|
| 112 |
+
"eval_steps_per_second": 0.237,
|
| 113 |
+
"step": 40
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 3.12,
|
| 117 |
+
"eval_loss": 0.26464152336120605,
|
| 118 |
+
"eval_runtime": 54.8467,
|
| 119 |
+
"eval_samples_per_second": 0.912,
|
| 120 |
+
"eval_steps_per_second": 0.237,
|
| 121 |
+
"step": 44
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 3.4,
|
| 125 |
+
"eval_loss": 0.25653430819511414,
|
| 126 |
+
"eval_runtime": 54.8327,
|
| 127 |
+
"eval_samples_per_second": 0.912,
|
| 128 |
+
"eval_steps_per_second": 0.237,
|
| 129 |
+
"step": 48
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 3.54,
|
| 133 |
+
"learning_rate": 5e-05,
|
| 134 |
+
"loss": 0.263,
|
| 135 |
+
"step": 50
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 3.68,
|
| 139 |
+
"eval_loss": 0.24627122282981873,
|
| 140 |
+
"eval_runtime": 54.813,
|
| 141 |
+
"eval_samples_per_second": 0.912,
|
| 142 |
+
"eval_steps_per_second": 0.237,
|
| 143 |
+
"step": 52
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 3.96,
|
| 147 |
+
"eval_loss": 0.23474617302417755,
|
| 148 |
+
"eval_runtime": 54.7901,
|
| 149 |
+
"eval_samples_per_second": 0.913,
|
| 150 |
+
"eval_steps_per_second": 0.237,
|
| 151 |
+
"step": 56
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 4.25,
|
| 155 |
+
"learning_rate": 6e-05,
|
| 156 |
+
"loss": 0.241,
|
| 157 |
+
"step": 60
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 4.25,
|
| 161 |
+
"eval_loss": 0.2220366895198822,
|
| 162 |
+
"eval_runtime": 54.7983,
|
| 163 |
+
"eval_samples_per_second": 0.912,
|
| 164 |
+
"eval_steps_per_second": 0.237,
|
| 165 |
+
"step": 60
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 4.53,
|
| 169 |
+
"eval_loss": 0.20926769077777863,
|
| 170 |
+
"eval_runtime": 54.7403,
|
| 171 |
+
"eval_samples_per_second": 0.913,
|
| 172 |
+
"eval_steps_per_second": 0.237,
|
| 173 |
+
"step": 64
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 4.81,
|
| 177 |
+
"eval_loss": 0.19629451632499695,
|
| 178 |
+
"eval_runtime": 54.7525,
|
| 179 |
+
"eval_samples_per_second": 0.913,
|
| 180 |
+
"eval_steps_per_second": 0.237,
|
| 181 |
+
"step": 68
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 4.96,
|
| 185 |
+
"learning_rate": 7e-05,
|
| 186 |
+
"loss": 0.2101,
|
| 187 |
+
"step": 70
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 5.1,
|
| 191 |
+
"eval_loss": 0.18524658679962158,
|
| 192 |
+
"eval_runtime": 54.7303,
|
| 193 |
+
"eval_samples_per_second": 0.914,
|
| 194 |
+
"eval_steps_per_second": 0.238,
|
| 195 |
+
"step": 72
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 5.38,
|
| 199 |
+
"eval_loss": 0.17731742560863495,
|
| 200 |
+
"eval_runtime": 54.7552,
|
| 201 |
+
"eval_samples_per_second": 0.913,
|
| 202 |
+
"eval_steps_per_second": 0.237,
|
| 203 |
+
"step": 76
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 5.66,
|
| 207 |
+
"learning_rate": 8e-05,
|
| 208 |
+
"loss": 0.1788,
|
| 209 |
+
"step": 80
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 5.66,
|
| 213 |
+
"eval_loss": 0.16993452608585358,
|
| 214 |
+
"eval_runtime": 54.729,
|
| 215 |
+
"eval_samples_per_second": 0.914,
|
| 216 |
+
"eval_steps_per_second": 0.238,
|
| 217 |
+
"step": 80
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 5.95,
|
| 221 |
+
"eval_loss": 0.164781853556633,
|
| 222 |
+
"eval_runtime": 54.741,
|
| 223 |
+
"eval_samples_per_second": 0.913,
|
| 224 |
+
"eval_steps_per_second": 0.237,
|
| 225 |
+
"step": 84
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 6.23,
|
| 229 |
+
"eval_loss": 0.16103117167949677,
|
| 230 |
+
"eval_runtime": 54.7837,
|
| 231 |
+
"eval_samples_per_second": 0.913,
|
| 232 |
+
"eval_steps_per_second": 0.237,
|
| 233 |
+
"step": 88
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 6.37,
|
| 237 |
+
"learning_rate": 9e-05,
|
| 238 |
+
"loss": 0.1615,
|
| 239 |
+
"step": 90
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 6.51,
|
| 243 |
+
"eval_loss": 0.15781742334365845,
|
| 244 |
+
"eval_runtime": 54.7138,
|
| 245 |
+
"eval_samples_per_second": 0.914,
|
| 246 |
+
"eval_steps_per_second": 0.238,
|
| 247 |
+
"step": 92
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 6.8,
|
| 251 |
+
"eval_loss": 0.15516981482505798,
|
| 252 |
+
"eval_runtime": 54.7516,
|
| 253 |
+
"eval_samples_per_second": 0.913,
|
| 254 |
+
"eval_steps_per_second": 0.237,
|
| 255 |
+
"step": 96
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 7.08,
|
| 259 |
+
"learning_rate": 0.0001,
|
| 260 |
+
"loss": 0.1533,
|
| 261 |
+
"step": 100
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 7.08,
|
| 265 |
+
"eval_loss": 0.15261690318584442,
|
| 266 |
+
"eval_runtime": 54.6891,
|
| 267 |
+
"eval_samples_per_second": 0.914,
|
| 268 |
+
"eval_steps_per_second": 0.238,
|
| 269 |
+
"step": 100
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 7.36,
|
| 273 |
+
"eval_loss": 0.15066812932491302,
|
| 274 |
+
"eval_runtime": 54.6884,
|
| 275 |
+
"eval_samples_per_second": 0.914,
|
| 276 |
+
"eval_steps_per_second": 0.238,
|
| 277 |
+
"step": 104
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 7.65,
|
| 281 |
+
"eval_loss": 0.14893724024295807,
|
| 282 |
+
"eval_runtime": 54.6275,
|
| 283 |
+
"eval_samples_per_second": 0.915,
|
| 284 |
+
"eval_steps_per_second": 0.238,
|
| 285 |
+
"step": 108
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 7.79,
|
| 289 |
+
"learning_rate": 9.090909090909092e-05,
|
| 290 |
+
"loss": 0.1463,
|
| 291 |
+
"step": 110
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 7.93,
|
| 295 |
+
"eval_loss": 0.14742153882980347,
|
| 296 |
+
"eval_runtime": 54.6174,
|
| 297 |
+
"eval_samples_per_second": 0.915,
|
| 298 |
+
"eval_steps_per_second": 0.238,
|
| 299 |
+
"step": 112
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 8.21,
|
| 303 |
+
"eval_loss": 0.14575307071208954,
|
| 304 |
+
"eval_runtime": 54.6366,
|
| 305 |
+
"eval_samples_per_second": 0.915,
|
| 306 |
+
"eval_steps_per_second": 0.238,
|
| 307 |
+
"step": 116
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 8.5,
|
| 311 |
+
"learning_rate": 8.181818181818183e-05,
|
| 312 |
+
"loss": 0.1399,
|
| 313 |
+
"step": 120
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 8.5,
|
| 317 |
+
"eval_loss": 0.14450186491012573,
|
| 318 |
+
"eval_runtime": 54.6303,
|
| 319 |
+
"eval_samples_per_second": 0.915,
|
| 320 |
+
"eval_steps_per_second": 0.238,
|
| 321 |
+
"step": 120
|
| 322 |
+
}
|
| 323 |
+
],
|
| 324 |
+
"logging_steps": 10,
|
| 325 |
+
"max_steps": 210,
|
| 326 |
+
"num_input_tokens_seen": 0,
|
| 327 |
+
"num_train_epochs": 15,
|
| 328 |
+
"save_steps": 12,
|
| 329 |
+
"total_flos": 3.9566860550445466e+17,
|
| 330 |
+
"train_batch_size": 4,
|
| 331 |
+
"trial_name": null,
|
| 332 |
+
"trial_params": null
|
| 333 |
+
}
|
checkpoint-120/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:85310c54a4f279d40e8badbc8f6f7406b57e15fc0c79500de525827feedf5072
|
| 3 |
+
size 4219
|
checkpoint-96/adapter_config.json
ADDED
|
@@ -0,0 +1,19 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"base_model_name_or_path": "codellama/CodeLlama-7b-hf",
|
| 3 |
+
"bias": "none",
|
| 4 |
+
"enable_lora": null,
|
| 5 |
+
"fan_in_fan_out": false,
|
| 6 |
+
"inference_mode": true,
|
| 7 |
+
"init_lora_weights": true,
|
| 8 |
+
"lora_alpha": 16,
|
| 9 |
+
"lora_dropout": 0.05,
|
| 10 |
+
"merge_weights": false,
|
| 11 |
+
"modules_to_save": null,
|
| 12 |
+
"peft_type": "LORA",
|
| 13 |
+
"r": 8,
|
| 14 |
+
"target_modules": [
|
| 15 |
+
"q_proj",
|
| 16 |
+
"v_proj"
|
| 17 |
+
],
|
| 18 |
+
"task_type": "CAUSAL_LM"
|
| 19 |
+
}
|
checkpoint-96/adapter_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0525137a8c5c2e3b8a67336c3faa14bb35cd8a82fe0e48fb5bbb1cc63517a683
|
| 3 |
+
size 16822989
|
checkpoint-96/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1b880941c0dde00e00c16127caaf28a86198b08be138268bce8178197dcf9531
|
| 3 |
+
size 33661637
|
checkpoint-96/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:989cc2f27d5146cf2ca96268118fce87275257fa4db48a24d9c45c7a01b6b870
|
| 3 |
+
size 14575
|
checkpoint-96/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:40465654f063e78a5631104698736f11b5438cb73c746c832c97195415dfdf51
|
| 3 |
+
size 627
|
checkpoint-96/trainer_state.json
ADDED
|
@@ -0,0 +1,267 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.15516981482505798,
|
| 3 |
+
"best_model_checkpoint": "./lora-out/checkpoint-96",
|
| 4 |
+
"epoch": 6.79646017699115,
|
| 5 |
+
"eval_steps": 4,
|
| 6 |
+
"global_step": 96,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.28,
|
| 13 |
+
"eval_loss": 0.3006412982940674,
|
| 14 |
+
"eval_runtime": 54.7333,
|
| 15 |
+
"eval_samples_per_second": 0.914,
|
| 16 |
+
"eval_steps_per_second": 0.238,
|
| 17 |
+
"step": 4
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.57,
|
| 21 |
+
"eval_loss": 0.300335168838501,
|
| 22 |
+
"eval_runtime": 54.8113,
|
| 23 |
+
"eval_samples_per_second": 0.912,
|
| 24 |
+
"eval_steps_per_second": 0.237,
|
| 25 |
+
"step": 8
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.71,
|
| 29 |
+
"learning_rate": 1e-05,
|
| 30 |
+
"loss": 0.3024,
|
| 31 |
+
"step": 10
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.85,
|
| 35 |
+
"eval_loss": 0.2993900179862976,
|
| 36 |
+
"eval_runtime": 54.782,
|
| 37 |
+
"eval_samples_per_second": 0.913,
|
| 38 |
+
"eval_steps_per_second": 0.237,
|
| 39 |
+
"step": 12
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 1.13,
|
| 43 |
+
"eval_loss": 0.29816487431526184,
|
| 44 |
+
"eval_runtime": 54.8049,
|
| 45 |
+
"eval_samples_per_second": 0.912,
|
| 46 |
+
"eval_steps_per_second": 0.237,
|
| 47 |
+
"step": 16
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 1.42,
|
| 51 |
+
"learning_rate": 2e-05,
|
| 52 |
+
"loss": 0.3035,
|
| 53 |
+
"step": 20
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.42,
|
| 57 |
+
"eval_loss": 0.29595255851745605,
|
| 58 |
+
"eval_runtime": 54.7885,
|
| 59 |
+
"eval_samples_per_second": 0.913,
|
| 60 |
+
"eval_steps_per_second": 0.237,
|
| 61 |
+
"step": 20
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 1.7,
|
| 65 |
+
"eval_loss": 0.2939557135105133,
|
| 66 |
+
"eval_runtime": 54.7999,
|
| 67 |
+
"eval_samples_per_second": 0.912,
|
| 68 |
+
"eval_steps_per_second": 0.237,
|
| 69 |
+
"step": 24
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 1.98,
|
| 73 |
+
"eval_loss": 0.29013773798942566,
|
| 74 |
+
"eval_runtime": 54.7805,
|
| 75 |
+
"eval_samples_per_second": 0.913,
|
| 76 |
+
"eval_steps_per_second": 0.237,
|
| 77 |
+
"step": 28
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 2.12,
|
| 81 |
+
"learning_rate": 3e-05,
|
| 82 |
+
"loss": 0.2959,
|
| 83 |
+
"step": 30
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 2.27,
|
| 87 |
+
"eval_loss": 0.28251081705093384,
|
| 88 |
+
"eval_runtime": 54.7706,
|
| 89 |
+
"eval_samples_per_second": 0.913,
|
| 90 |
+
"eval_steps_per_second": 0.237,
|
| 91 |
+
"step": 32
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 2.55,
|
| 95 |
+
"eval_loss": 0.2771329879760742,
|
| 96 |
+
"eval_runtime": 54.7818,
|
| 97 |
+
"eval_samples_per_second": 0.913,
|
| 98 |
+
"eval_steps_per_second": 0.237,
|
| 99 |
+
"step": 36
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 2.83,
|
| 103 |
+
"learning_rate": 4e-05,
|
| 104 |
+
"loss": 0.284,
|
| 105 |
+
"step": 40
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 2.83,
|
| 109 |
+
"eval_loss": 0.27146488428115845,
|
| 110 |
+
"eval_runtime": 54.803,
|
| 111 |
+
"eval_samples_per_second": 0.912,
|
| 112 |
+
"eval_steps_per_second": 0.237,
|
| 113 |
+
"step": 40
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 3.12,
|
| 117 |
+
"eval_loss": 0.26464152336120605,
|
| 118 |
+
"eval_runtime": 54.8467,
|
| 119 |
+
"eval_samples_per_second": 0.912,
|
| 120 |
+
"eval_steps_per_second": 0.237,
|
| 121 |
+
"step": 44
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 3.4,
|
| 125 |
+
"eval_loss": 0.25653430819511414,
|
| 126 |
+
"eval_runtime": 54.8327,
|
| 127 |
+
"eval_samples_per_second": 0.912,
|
| 128 |
+
"eval_steps_per_second": 0.237,
|
| 129 |
+
"step": 48
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 3.54,
|
| 133 |
+
"learning_rate": 5e-05,
|
| 134 |
+
"loss": 0.263,
|
| 135 |
+
"step": 50
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 3.68,
|
| 139 |
+
"eval_loss": 0.24627122282981873,
|
| 140 |
+
"eval_runtime": 54.813,
|
| 141 |
+
"eval_samples_per_second": 0.912,
|
| 142 |
+
"eval_steps_per_second": 0.237,
|
| 143 |
+
"step": 52
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 3.96,
|
| 147 |
+
"eval_loss": 0.23474617302417755,
|
| 148 |
+
"eval_runtime": 54.7901,
|
| 149 |
+
"eval_samples_per_second": 0.913,
|
| 150 |
+
"eval_steps_per_second": 0.237,
|
| 151 |
+
"step": 56
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 4.25,
|
| 155 |
+
"learning_rate": 6e-05,
|
| 156 |
+
"loss": 0.241,
|
| 157 |
+
"step": 60
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 4.25,
|
| 161 |
+
"eval_loss": 0.2220366895198822,
|
| 162 |
+
"eval_runtime": 54.7983,
|
| 163 |
+
"eval_samples_per_second": 0.912,
|
| 164 |
+
"eval_steps_per_second": 0.237,
|
| 165 |
+
"step": 60
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 4.53,
|
| 169 |
+
"eval_loss": 0.20926769077777863,
|
| 170 |
+
"eval_runtime": 54.7403,
|
| 171 |
+
"eval_samples_per_second": 0.913,
|
| 172 |
+
"eval_steps_per_second": 0.237,
|
| 173 |
+
"step": 64
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 4.81,
|
| 177 |
+
"eval_loss": 0.19629451632499695,
|
| 178 |
+
"eval_runtime": 54.7525,
|
| 179 |
+
"eval_samples_per_second": 0.913,
|
| 180 |
+
"eval_steps_per_second": 0.237,
|
| 181 |
+
"step": 68
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 4.96,
|
| 185 |
+
"learning_rate": 7e-05,
|
| 186 |
+
"loss": 0.2101,
|
| 187 |
+
"step": 70
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 5.1,
|
| 191 |
+
"eval_loss": 0.18524658679962158,
|
| 192 |
+
"eval_runtime": 54.7303,
|
| 193 |
+
"eval_samples_per_second": 0.914,
|
| 194 |
+
"eval_steps_per_second": 0.238,
|
| 195 |
+
"step": 72
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 5.38,
|
| 199 |
+
"eval_loss": 0.17731742560863495,
|
| 200 |
+
"eval_runtime": 54.7552,
|
| 201 |
+
"eval_samples_per_second": 0.913,
|
| 202 |
+
"eval_steps_per_second": 0.237,
|
| 203 |
+
"step": 76
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 5.66,
|
| 207 |
+
"learning_rate": 8e-05,
|
| 208 |
+
"loss": 0.1788,
|
| 209 |
+
"step": 80
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 5.66,
|
| 213 |
+
"eval_loss": 0.16993452608585358,
|
| 214 |
+
"eval_runtime": 54.729,
|
| 215 |
+
"eval_samples_per_second": 0.914,
|
| 216 |
+
"eval_steps_per_second": 0.238,
|
| 217 |
+
"step": 80
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 5.95,
|
| 221 |
+
"eval_loss": 0.164781853556633,
|
| 222 |
+
"eval_runtime": 54.741,
|
| 223 |
+
"eval_samples_per_second": 0.913,
|
| 224 |
+
"eval_steps_per_second": 0.237,
|
| 225 |
+
"step": 84
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 6.23,
|
| 229 |
+
"eval_loss": 0.16103117167949677,
|
| 230 |
+
"eval_runtime": 54.7837,
|
| 231 |
+
"eval_samples_per_second": 0.913,
|
| 232 |
+
"eval_steps_per_second": 0.237,
|
| 233 |
+
"step": 88
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 6.37,
|
| 237 |
+
"learning_rate": 9e-05,
|
| 238 |
+
"loss": 0.1615,
|
| 239 |
+
"step": 90
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 6.51,
|
| 243 |
+
"eval_loss": 0.15781742334365845,
|
| 244 |
+
"eval_runtime": 54.7138,
|
| 245 |
+
"eval_samples_per_second": 0.914,
|
| 246 |
+
"eval_steps_per_second": 0.238,
|
| 247 |
+
"step": 92
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 6.8,
|
| 251 |
+
"eval_loss": 0.15516981482505798,
|
| 252 |
+
"eval_runtime": 54.7516,
|
| 253 |
+
"eval_samples_per_second": 0.913,
|
| 254 |
+
"eval_steps_per_second": 0.237,
|
| 255 |
+
"step": 96
|
| 256 |
+
}
|
| 257 |
+
],
|
| 258 |
+
"logging_steps": 10,
|
| 259 |
+
"max_steps": 210,
|
| 260 |
+
"num_input_tokens_seen": 0,
|
| 261 |
+
"num_train_epochs": 15,
|
| 262 |
+
"save_steps": 12,
|
| 263 |
+
"total_flos": 3.166594152769782e+17,
|
| 264 |
+
"train_batch_size": 4,
|
| 265 |
+
"trial_name": null,
|
| 266 |
+
"trial_params": null
|
| 267 |
+
}
|
checkpoint-96/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:85310c54a4f279d40e8badbc8f6f7406b57e15fc0c79500de525827feedf5072
|
| 3 |
+
size 4219
|