DioulaD/classificateur-intention_camembert
Browse files- all_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +190 -0
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 20.0,
|
| 3 |
+
"train_loss": 0.11865015212446452,
|
| 4 |
+
"train_runtime": 23.7587,
|
| 5 |
+
"train_samples": 72,
|
| 6 |
+
"train_samples_per_second": 67.344,
|
| 7 |
+
"train_steps_per_second": 4.209
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 20.0,
|
| 3 |
+
"train_loss": 0.11865015212446452,
|
| 4 |
+
"train_runtime": 23.7587,
|
| 5 |
+
"train_samples": 72,
|
| 6 |
+
"train_samples_per_second": 67.344,
|
| 7 |
+
"train_steps_per_second": 4.209
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,190 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 20.0,
|
| 5 |
+
"eval_steps": 10,
|
| 6 |
+
"global_step": 100,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 2.0,
|
| 13 |
+
"grad_norm": 2.5965750217437744,
|
| 14 |
+
"learning_rate": 0.00018,
|
| 15 |
+
"loss": 0.7517,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 2.0,
|
| 20 |
+
"eval_accuracy": 0.8888888888888888,
|
| 21 |
+
"eval_loss": 0.4700092077255249,
|
| 22 |
+
"eval_runtime": 0.0309,
|
| 23 |
+
"eval_samples_per_second": 582.425,
|
| 24 |
+
"eval_steps_per_second": 64.714,
|
| 25 |
+
"step": 10
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 4.0,
|
| 29 |
+
"grad_norm": 0.535392701625824,
|
| 30 |
+
"learning_rate": 0.00016,
|
| 31 |
+
"loss": 0.2834,
|
| 32 |
+
"step": 20
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 4.0,
|
| 36 |
+
"eval_accuracy": 0.8888888888888888,
|
| 37 |
+
"eval_loss": 0.3312886953353882,
|
| 38 |
+
"eval_runtime": 0.0245,
|
| 39 |
+
"eval_samples_per_second": 735.112,
|
| 40 |
+
"eval_steps_per_second": 81.679,
|
| 41 |
+
"step": 20
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 6.0,
|
| 45 |
+
"grad_norm": 0.14222599565982819,
|
| 46 |
+
"learning_rate": 0.00014,
|
| 47 |
+
"loss": 0.0488,
|
| 48 |
+
"step": 30
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 6.0,
|
| 52 |
+
"eval_accuracy": 0.8888888888888888,
|
| 53 |
+
"eval_loss": 0.35279521346092224,
|
| 54 |
+
"eval_runtime": 0.0265,
|
| 55 |
+
"eval_samples_per_second": 678.105,
|
| 56 |
+
"eval_steps_per_second": 75.345,
|
| 57 |
+
"step": 30
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 8.0,
|
| 61 |
+
"grad_norm": 0.09212491661310196,
|
| 62 |
+
"learning_rate": 0.00012,
|
| 63 |
+
"loss": 0.0181,
|
| 64 |
+
"step": 40
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 8.0,
|
| 68 |
+
"eval_accuracy": 0.8888888888888888,
|
| 69 |
+
"eval_loss": 0.6354701519012451,
|
| 70 |
+
"eval_runtime": 0.0268,
|
| 71 |
+
"eval_samples_per_second": 672.847,
|
| 72 |
+
"eval_steps_per_second": 74.761,
|
| 73 |
+
"step": 40
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 10.0,
|
| 77 |
+
"grad_norm": 0.05534437298774719,
|
| 78 |
+
"learning_rate": 0.0001,
|
| 79 |
+
"loss": 0.0079,
|
| 80 |
+
"step": 50
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 10.0,
|
| 84 |
+
"eval_accuracy": 0.8888888888888888,
|
| 85 |
+
"eval_loss": 0.6675615310668945,
|
| 86 |
+
"eval_runtime": 0.0382,
|
| 87 |
+
"eval_samples_per_second": 471.491,
|
| 88 |
+
"eval_steps_per_second": 52.388,
|
| 89 |
+
"step": 50
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 12.0,
|
| 93 |
+
"grad_norm": 0.04630253463983536,
|
| 94 |
+
"learning_rate": 8e-05,
|
| 95 |
+
"loss": 0.0437,
|
| 96 |
+
"step": 60
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 12.0,
|
| 100 |
+
"eval_accuracy": 0.8888888888888888,
|
| 101 |
+
"eval_loss": 0.5816599130630493,
|
| 102 |
+
"eval_runtime": 0.0247,
|
| 103 |
+
"eval_samples_per_second": 728.213,
|
| 104 |
+
"eval_steps_per_second": 80.913,
|
| 105 |
+
"step": 60
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 14.0,
|
| 109 |
+
"grad_norm": 0.04637857526540756,
|
| 110 |
+
"learning_rate": 6e-05,
|
| 111 |
+
"loss": 0.0049,
|
| 112 |
+
"step": 70
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 14.0,
|
| 116 |
+
"eval_accuracy": 0.8888888888888888,
|
| 117 |
+
"eval_loss": 0.4498683512210846,
|
| 118 |
+
"eval_runtime": 0.0252,
|
| 119 |
+
"eval_samples_per_second": 713.378,
|
| 120 |
+
"eval_steps_per_second": 79.264,
|
| 121 |
+
"step": 70
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 16.0,
|
| 125 |
+
"grad_norm": 0.1237090453505516,
|
| 126 |
+
"learning_rate": 4e-05,
|
| 127 |
+
"loss": 0.0192,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 16.0,
|
| 132 |
+
"eval_accuracy": 0.8888888888888888,
|
| 133 |
+
"eval_loss": 0.516182541847229,
|
| 134 |
+
"eval_runtime": 0.0258,
|
| 135 |
+
"eval_samples_per_second": 698.146,
|
| 136 |
+
"eval_steps_per_second": 77.572,
|
| 137 |
+
"step": 80
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 18.0,
|
| 141 |
+
"grad_norm": 0.03347891941666603,
|
| 142 |
+
"learning_rate": 2e-05,
|
| 143 |
+
"loss": 0.0045,
|
| 144 |
+
"step": 90
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 18.0,
|
| 148 |
+
"eval_accuracy": 0.8888888888888888,
|
| 149 |
+
"eval_loss": 0.5420387983322144,
|
| 150 |
+
"eval_runtime": 0.0267,
|
| 151 |
+
"eval_samples_per_second": 673.549,
|
| 152 |
+
"eval_steps_per_second": 74.839,
|
| 153 |
+
"step": 90
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 20.0,
|
| 157 |
+
"grad_norm": 0.03203904256224632,
|
| 158 |
+
"learning_rate": 0.0,
|
| 159 |
+
"loss": 0.0042,
|
| 160 |
+
"step": 100
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 20.0,
|
| 164 |
+
"eval_accuracy": 0.8888888888888888,
|
| 165 |
+
"eval_loss": 0.5463488101959229,
|
| 166 |
+
"eval_runtime": 0.0257,
|
| 167 |
+
"eval_samples_per_second": 699.731,
|
| 168 |
+
"eval_steps_per_second": 77.748,
|
| 169 |
+
"step": 100
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 20.0,
|
| 173 |
+
"step": 100,
|
| 174 |
+
"total_flos": 16662479407056.0,
|
| 175 |
+
"train_loss": 0.11865015212446452,
|
| 176 |
+
"train_runtime": 23.7587,
|
| 177 |
+
"train_samples_per_second": 67.344,
|
| 178 |
+
"train_steps_per_second": 4.209
|
| 179 |
+
}
|
| 180 |
+
],
|
| 181 |
+
"logging_steps": 10,
|
| 182 |
+
"max_steps": 100,
|
| 183 |
+
"num_input_tokens_seen": 0,
|
| 184 |
+
"num_train_epochs": 20,
|
| 185 |
+
"save_steps": 20,
|
| 186 |
+
"total_flos": 16662479407056.0,
|
| 187 |
+
"train_batch_size": 16,
|
| 188 |
+
"trial_name": null,
|
| 189 |
+
"trial_params": null
|
| 190 |
+
}
|