starlineventures/pilot-talk
Browse files- README.md +1 -1
- adapter_config.json +4 -4
- adapter_model.safetensors +1 -1
- all_results.json +5 -5
- train_results.json +5 -5
- trainer_state.json +74 -172
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -36,7 +36,7 @@ More information needed
|
|
| 36 |
|
| 37 |
The following hyperparameters were used during training:
|
| 38 |
- learning_rate: 0.0001
|
| 39 |
-
- train_batch_size:
|
| 40 |
- eval_batch_size: 16
|
| 41 |
- seed: 3407
|
| 42 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
|
|
|
| 36 |
|
| 37 |
The following hyperparameters were used during training:
|
| 38 |
- learning_rate: 0.0001
|
| 39 |
+
- train_batch_size: 5
|
| 40 |
- eval_batch_size: 16
|
| 41 |
- seed: 3407
|
| 42 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
adapter_config.json
CHANGED
|
@@ -23,13 +23,13 @@
|
|
| 23 |
"rank_pattern": {},
|
| 24 |
"revision": null,
|
| 25 |
"target_modules": [
|
| 26 |
-
"
|
| 27 |
"down_proj",
|
| 28 |
-
"o_proj",
|
| 29 |
-
"q_proj",
|
| 30 |
"v_proj",
|
| 31 |
"up_proj",
|
| 32 |
-
"
|
|
|
|
|
|
|
| 33 |
],
|
| 34 |
"task_type": null,
|
| 35 |
"use_dora": false,
|
|
|
|
| 23 |
"rank_pattern": {},
|
| 24 |
"revision": null,
|
| 25 |
"target_modules": [
|
| 26 |
+
"k_proj",
|
| 27 |
"down_proj",
|
|
|
|
|
|
|
| 28 |
"v_proj",
|
| 29 |
"up_proj",
|
| 30 |
+
"o_proj",
|
| 31 |
+
"q_proj",
|
| 32 |
+
"gate_proj"
|
| 33 |
],
|
| 34 |
"task_type": null,
|
| 35 |
"use_dora": false,
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 94422368
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:54dba78f4b6e627f89fc3f21c2541ef05ddf6bb95172877a10b7ba7e8f292741
|
| 3 |
size 94422368
|
all_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
"total_flos": 0.0,
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
"total_flos": 0.0,
|
| 4 |
+
"train_loss": 0.30142025277018547,
|
| 5 |
+
"train_runtime": 98.1187,
|
| 6 |
+
"train_samples_per_second": 24.46,
|
| 7 |
+
"train_steps_per_second": 4.892
|
| 8 |
}
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
"total_flos": 0.0,
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
"total_flos": 0.0,
|
| 4 |
+
"train_loss": 0.30142025277018547,
|
| 5 |
+
"train_runtime": 98.1187,
|
| 6 |
+
"train_samples_per_second": 24.46,
|
| 7 |
+
"train_steps_per_second": 4.892
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,235 +1,137 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 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.
|
| 13 |
-
"grad_norm":
|
| 14 |
-
"learning_rate": 9.
|
| 15 |
-
"loss":
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm":
|
| 21 |
-
"learning_rate": 9.
|
| 22 |
-
"loss": 0.
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch": 0.
|
| 27 |
-
"grad_norm":
|
| 28 |
-
"learning_rate": 9.
|
| 29 |
-
"loss": 0.
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"grad_norm":
|
| 35 |
-
"learning_rate":
|
| 36 |
-
"loss":
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
-
"epoch": 0.
|
| 41 |
-
"grad_norm":
|
| 42 |
-
"learning_rate": 8.
|
| 43 |
-
"loss": 0.
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"grad_norm":
|
| 49 |
-
"learning_rate": 8.
|
| 50 |
-
"loss": 0.
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"grad_norm":
|
| 56 |
-
"learning_rate":
|
| 57 |
-
"loss": 0.
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
-
"epoch": 0.
|
| 62 |
-
"grad_norm":
|
| 63 |
-
"learning_rate":
|
| 64 |
-
"loss": 0.
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
-
"epoch": 0.
|
| 69 |
-
"grad_norm":
|
| 70 |
-
"learning_rate":
|
| 71 |
-
"loss": 0.
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
-
"epoch": 0.
|
| 76 |
-
"grad_norm":
|
| 77 |
-
"learning_rate":
|
| 78 |
-
"loss": 0.
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
-
"epoch":
|
| 83 |
-
"grad_norm":
|
| 84 |
-
"learning_rate":
|
| 85 |
-
"loss": 0.
|
| 86 |
"step": 110
|
| 87 |
},
|
| 88 |
{
|
| 89 |
-
"epoch":
|
| 90 |
-
"grad_norm":
|
| 91 |
-
"learning_rate":
|
| 92 |
-
"loss": 0.
|
| 93 |
"step": 120
|
| 94 |
},
|
| 95 |
{
|
| 96 |
-
"epoch":
|
| 97 |
-
"grad_norm":
|
| 98 |
-
"learning_rate":
|
| 99 |
-
"loss": 0.
|
| 100 |
"step": 130
|
| 101 |
},
|
| 102 |
{
|
| 103 |
-
"epoch":
|
| 104 |
-
"grad_norm":
|
| 105 |
-
"learning_rate":
|
| 106 |
-
"loss": 0.
|
| 107 |
"step": 140
|
| 108 |
},
|
| 109 |
{
|
| 110 |
-
"epoch":
|
| 111 |
-
"grad_norm":
|
| 112 |
-
"learning_rate":
|
| 113 |
-
"loss": 0.
|
| 114 |
"step": 150
|
| 115 |
},
|
| 116 |
{
|
| 117 |
-
"epoch": 1.
|
| 118 |
-
"grad_norm":
|
| 119 |
-
"learning_rate":
|
| 120 |
-
"loss": 0.
|
| 121 |
"step": 160
|
| 122 |
},
|
| 123 |
{
|
| 124 |
-
"epoch": 1.
|
| 125 |
-
"
|
| 126 |
-
"learning_rate": 4.4444444444444447e-05,
|
| 127 |
-
"loss": 0.4554,
|
| 128 |
-
"step": 170
|
| 129 |
-
},
|
| 130 |
-
{
|
| 131 |
-
"epoch": 1.7647058823529411,
|
| 132 |
-
"grad_norm": 23.189788818359375,
|
| 133 |
-
"learning_rate": 4.11764705882353e-05,
|
| 134 |
-
"loss": 0.6165,
|
| 135 |
-
"step": 180
|
| 136 |
-
},
|
| 137 |
-
{
|
| 138 |
-
"epoch": 1.8627450980392157,
|
| 139 |
-
"grad_norm": 21.400800704956055,
|
| 140 |
-
"learning_rate": 3.790849673202614e-05,
|
| 141 |
-
"loss": 0.3971,
|
| 142 |
-
"step": 190
|
| 143 |
-
},
|
| 144 |
-
{
|
| 145 |
-
"epoch": 1.9607843137254903,
|
| 146 |
-
"grad_norm": 20.903608322143555,
|
| 147 |
-
"learning_rate": 3.464052287581699e-05,
|
| 148 |
-
"loss": 0.5734,
|
| 149 |
-
"step": 200
|
| 150 |
-
},
|
| 151 |
-
{
|
| 152 |
-
"epoch": 2.0588235294117645,
|
| 153 |
-
"grad_norm": 23.240053176879883,
|
| 154 |
-
"learning_rate": 3.137254901960784e-05,
|
| 155 |
-
"loss": 0.4406,
|
| 156 |
-
"step": 210
|
| 157 |
-
},
|
| 158 |
-
{
|
| 159 |
-
"epoch": 2.156862745098039,
|
| 160 |
-
"grad_norm": 24.06924057006836,
|
| 161 |
-
"learning_rate": 2.8104575163398693e-05,
|
| 162 |
-
"loss": 0.4119,
|
| 163 |
-
"step": 220
|
| 164 |
-
},
|
| 165 |
-
{
|
| 166 |
-
"epoch": 2.2549019607843137,
|
| 167 |
-
"grad_norm": 24.296913146972656,
|
| 168 |
-
"learning_rate": 2.4836601307189544e-05,
|
| 169 |
-
"loss": 0.5279,
|
| 170 |
-
"step": 230
|
| 171 |
-
},
|
| 172 |
-
{
|
| 173 |
-
"epoch": 2.3529411764705883,
|
| 174 |
-
"grad_norm": 20.898887634277344,
|
| 175 |
-
"learning_rate": 2.1568627450980395e-05,
|
| 176 |
-
"loss": 0.5091,
|
| 177 |
-
"step": 240
|
| 178 |
-
},
|
| 179 |
-
{
|
| 180 |
-
"epoch": 2.450980392156863,
|
| 181 |
-
"grad_norm": 22.029281616210938,
|
| 182 |
-
"learning_rate": 1.8300653594771242e-05,
|
| 183 |
-
"loss": 0.4114,
|
| 184 |
-
"step": 250
|
| 185 |
-
},
|
| 186 |
-
{
|
| 187 |
-
"epoch": 2.549019607843137,
|
| 188 |
-
"grad_norm": 23.726993560791016,
|
| 189 |
-
"learning_rate": 1.5032679738562091e-05,
|
| 190 |
-
"loss": 0.4127,
|
| 191 |
-
"step": 260
|
| 192 |
-
},
|
| 193 |
-
{
|
| 194 |
-
"epoch": 2.6470588235294117,
|
| 195 |
-
"grad_norm": 20.965084075927734,
|
| 196 |
-
"learning_rate": 1.1764705882352942e-05,
|
| 197 |
-
"loss": 0.4927,
|
| 198 |
-
"step": 270
|
| 199 |
-
},
|
| 200 |
-
{
|
| 201 |
-
"epoch": 2.7450980392156863,
|
| 202 |
-
"grad_norm": 21.0610294342041,
|
| 203 |
-
"learning_rate": 8.496732026143791e-06,
|
| 204 |
-
"loss": 0.5405,
|
| 205 |
-
"step": 280
|
| 206 |
-
},
|
| 207 |
-
{
|
| 208 |
-
"epoch": 2.843137254901961,
|
| 209 |
-
"grad_norm": 20.865800857543945,
|
| 210 |
-
"learning_rate": 5.228758169934641e-06,
|
| 211 |
-
"loss": 0.4165,
|
| 212 |
-
"step": 290
|
| 213 |
-
},
|
| 214 |
-
{
|
| 215 |
-
"epoch": 2.9411764705882355,
|
| 216 |
-
"grad_norm": 21.134117126464844,
|
| 217 |
-
"learning_rate": 1.96078431372549e-06,
|
| 218 |
-
"loss": 0.4396,
|
| 219 |
-
"step": 300
|
| 220 |
-
},
|
| 221 |
-
{
|
| 222 |
-
"epoch": 3.0,
|
| 223 |
-
"step": 306,
|
| 224 |
"total_flos": 0.0,
|
| 225 |
-
"train_loss": 0.
|
| 226 |
-
"train_runtime":
|
| 227 |
-
"train_samples_per_second":
|
| 228 |
-
"train_steps_per_second":
|
| 229 |
}
|
| 230 |
],
|
| 231 |
"logging_steps": 10,
|
| 232 |
-
"max_steps":
|
| 233 |
"num_input_tokens_seen": 0,
|
| 234 |
"num_train_epochs": 3,
|
| 235 |
"save_steps": 500,
|
|
@@ -246,7 +148,7 @@
|
|
| 246 |
}
|
| 247 |
},
|
| 248 |
"total_flos": 0.0,
|
| 249 |
-
"train_batch_size":
|
| 250 |
"trial_name": null,
|
| 251 |
"trial_params": null
|
| 252 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 160,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.0625,
|
| 13 |
+
"grad_norm": 20.87160873413086,
|
| 14 |
+
"learning_rate": 9.791666666666667e-05,
|
| 15 |
+
"loss": 1.7149,
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.125,
|
| 20 |
+
"grad_norm": 21.004980087280273,
|
| 21 |
+
"learning_rate": 9.583333333333334e-05,
|
| 22 |
+
"loss": 0.8011,
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.1875,
|
| 27 |
+
"grad_norm": 20.969247817993164,
|
| 28 |
+
"learning_rate": 9.375e-05,
|
| 29 |
+
"loss": 0.3873,
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.25,
|
| 34 |
+
"grad_norm": 23.872455596923828,
|
| 35 |
+
"learning_rate": 9.166666666666667e-05,
|
| 36 |
+
"loss": 0.2433,
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.3125,
|
| 41 |
+
"grad_norm": 26.21103286743164,
|
| 42 |
+
"learning_rate": 8.958333333333335e-05,
|
| 43 |
+
"loss": 0.1846,
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.375,
|
| 48 |
+
"grad_norm": 30.174484252929688,
|
| 49 |
+
"learning_rate": 8.75e-05,
|
| 50 |
+
"loss": 0.1698,
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.4375,
|
| 55 |
+
"grad_norm": 30.354196548461914,
|
| 56 |
+
"learning_rate": 8.541666666666666e-05,
|
| 57 |
+
"loss": 0.1519,
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 0.5,
|
| 62 |
+
"grad_norm": 30.821664810180664,
|
| 63 |
+
"learning_rate": 8.333333333333334e-05,
|
| 64 |
+
"loss": 0.143,
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 0.5625,
|
| 69 |
+
"grad_norm": 35.24763488769531,
|
| 70 |
+
"learning_rate": 8.125000000000001e-05,
|
| 71 |
+
"loss": 0.1373,
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
+
"epoch": 0.625,
|
| 76 |
+
"grad_norm": 34.299808502197266,
|
| 77 |
+
"learning_rate": 7.916666666666666e-05,
|
| 78 |
+
"loss": 0.1344,
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
+
"epoch": 0.6875,
|
| 83 |
+
"grad_norm": 32.865936279296875,
|
| 84 |
+
"learning_rate": 7.708333333333334e-05,
|
| 85 |
+
"loss": 0.1282,
|
| 86 |
"step": 110
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 0.75,
|
| 90 |
+
"grad_norm": 31.330398559570312,
|
| 91 |
+
"learning_rate": 7.500000000000001e-05,
|
| 92 |
+
"loss": 0.1282,
|
| 93 |
"step": 120
|
| 94 |
},
|
| 95 |
{
|
| 96 |
+
"epoch": 0.8125,
|
| 97 |
+
"grad_norm": 30.16261100769043,
|
| 98 |
+
"learning_rate": 7.291666666666667e-05,
|
| 99 |
+
"loss": 0.1232,
|
| 100 |
"step": 130
|
| 101 |
},
|
| 102 |
{
|
| 103 |
+
"epoch": 0.875,
|
| 104 |
+
"grad_norm": 29.665931701660156,
|
| 105 |
+
"learning_rate": 7.083333333333334e-05,
|
| 106 |
+
"loss": 0.1261,
|
| 107 |
"step": 140
|
| 108 |
},
|
| 109 |
{
|
| 110 |
+
"epoch": 0.9375,
|
| 111 |
+
"grad_norm": 30.841318130493164,
|
| 112 |
+
"learning_rate": 6.875e-05,
|
| 113 |
+
"loss": 0.1241,
|
| 114 |
"step": 150
|
| 115 |
},
|
| 116 |
{
|
| 117 |
+
"epoch": 1.0,
|
| 118 |
+
"grad_norm": 32.89496994018555,
|
| 119 |
+
"learning_rate": 6.666666666666667e-05,
|
| 120 |
+
"loss": 0.1253,
|
| 121 |
"step": 160
|
| 122 |
},
|
| 123 |
{
|
| 124 |
+
"epoch": 1.0,
|
| 125 |
+
"step": 160,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 126 |
"total_flos": 0.0,
|
| 127 |
+
"train_loss": 0.30142025277018547,
|
| 128 |
+
"train_runtime": 98.1187,
|
| 129 |
+
"train_samples_per_second": 24.46,
|
| 130 |
+
"train_steps_per_second": 4.892
|
| 131 |
}
|
| 132 |
],
|
| 133 |
"logging_steps": 10,
|
| 134 |
+
"max_steps": 480,
|
| 135 |
"num_input_tokens_seen": 0,
|
| 136 |
"num_train_epochs": 3,
|
| 137 |
"save_steps": 500,
|
|
|
|
| 148 |
}
|
| 149 |
},
|
| 150 |
"total_flos": 0.0,
|
| 151 |
+
"train_batch_size": 5,
|
| 152 |
"trial_name": null,
|
| 153 |
"trial_params": null
|
| 154 |
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5432
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7c4f9fc9bff2e218da5e9f8f7b396e54282341ce6e9263a7f1e576e953a84151
|
| 3 |
size 5432
|