DioulaD/classifier-camembert
Browse files- all_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +94 -0
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 80.0,
|
| 3 |
+
"train_loss": 0.03407109707593918,
|
| 4 |
+
"train_runtime": 41.4151,
|
| 5 |
+
"train_samples": 72,
|
| 6 |
+
"train_samples_per_second": 154.533,
|
| 7 |
+
"train_steps_per_second": 9.658
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 80.0,
|
| 3 |
+
"train_loss": 0.03407109707593918,
|
| 4 |
+
"train_runtime": 41.4151,
|
| 5 |
+
"train_samples": 72,
|
| 6 |
+
"train_samples_per_second": 154.533,
|
| 7 |
+
"train_steps_per_second": 9.658
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,94 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 80.0,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 400,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 20.0,
|
| 13 |
+
"grad_norm": 0.009852898307144642,
|
| 14 |
+
"learning_rate": 0.00015000000000000001,
|
| 15 |
+
"loss": 0.1022,
|
| 16 |
+
"step": 100
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 20.0,
|
| 20 |
+
"eval_accuracy": 1.0,
|
| 21 |
+
"eval_loss": 0.0008228308288380504,
|
| 22 |
+
"eval_runtime": 0.0347,
|
| 23 |
+
"eval_samples_per_second": 519.229,
|
| 24 |
+
"eval_steps_per_second": 57.692,
|
| 25 |
+
"step": 100
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 40.0,
|
| 29 |
+
"grad_norm": 0.008125479333102703,
|
| 30 |
+
"learning_rate": 0.0001,
|
| 31 |
+
"loss": 0.006,
|
| 32 |
+
"step": 200
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 40.0,
|
| 36 |
+
"eval_accuracy": 1.0,
|
| 37 |
+
"eval_loss": 0.0006637517362833023,
|
| 38 |
+
"eval_runtime": 0.028,
|
| 39 |
+
"eval_samples_per_second": 642.488,
|
| 40 |
+
"eval_steps_per_second": 71.388,
|
| 41 |
+
"step": 200
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 60.0,
|
| 45 |
+
"grad_norm": 0.00710184033960104,
|
| 46 |
+
"learning_rate": 5e-05,
|
| 47 |
+
"loss": 0.0223,
|
| 48 |
+
"step": 300
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 60.0,
|
| 52 |
+
"eval_accuracy": 1.0,
|
| 53 |
+
"eval_loss": 0.0019779244903475046,
|
| 54 |
+
"eval_runtime": 0.0323,
|
| 55 |
+
"eval_samples_per_second": 557.646,
|
| 56 |
+
"eval_steps_per_second": 61.961,
|
| 57 |
+
"step": 300
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 80.0,
|
| 61 |
+
"grad_norm": 0.00847003422677517,
|
| 62 |
+
"learning_rate": 0.0,
|
| 63 |
+
"loss": 0.0058,
|
| 64 |
+
"step": 400
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 80.0,
|
| 68 |
+
"eval_accuracy": 1.0,
|
| 69 |
+
"eval_loss": 0.0008943404536694288,
|
| 70 |
+
"eval_runtime": 0.032,
|
| 71 |
+
"eval_samples_per_second": 562.516,
|
| 72 |
+
"eval_steps_per_second": 62.502,
|
| 73 |
+
"step": 400
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 80.0,
|
| 77 |
+
"step": 400,
|
| 78 |
+
"total_flos": 66867808427280.0,
|
| 79 |
+
"train_loss": 0.03407109707593918,
|
| 80 |
+
"train_runtime": 41.4151,
|
| 81 |
+
"train_samples_per_second": 154.533,
|
| 82 |
+
"train_steps_per_second": 9.658
|
| 83 |
+
}
|
| 84 |
+
],
|
| 85 |
+
"logging_steps": 100,
|
| 86 |
+
"max_steps": 400,
|
| 87 |
+
"num_input_tokens_seen": 0,
|
| 88 |
+
"num_train_epochs": 80,
|
| 89 |
+
"save_steps": 100,
|
| 90 |
+
"total_flos": 66867808427280.0,
|
| 91 |
+
"train_batch_size": 16,
|
| 92 |
+
"trial_name": null,
|
| 93 |
+
"trial_params": null
|
| 94 |
+
}
|