Upload 8 files
Browse files- config.json +33 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- preprocessor_config.json +22 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +885 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "google/vit-base-patch16-224",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"CustomViTForImageClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.2,
|
| 7 |
+
"encoder_stride": 16,
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.2,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"id2label": {
|
| 12 |
+
"0": "benigno",
|
| 13 |
+
"1": "maligno",
|
| 14 |
+
"2": "sospechoso"
|
| 15 |
+
},
|
| 16 |
+
"image_size": 224,
|
| 17 |
+
"initializer_range": 0.02,
|
| 18 |
+
"intermediate_size": 3072,
|
| 19 |
+
"label2id": {
|
| 20 |
+
"benigno": 0,
|
| 21 |
+
"maligno": 1,
|
| 22 |
+
"sospechoso": 2
|
| 23 |
+
},
|
| 24 |
+
"layer_norm_eps": 1e-12,
|
| 25 |
+
"model_type": "vit",
|
| 26 |
+
"num_attention_heads": 12,
|
| 27 |
+
"num_channels": 3,
|
| 28 |
+
"num_hidden_layers": 12,
|
| 29 |
+
"patch_size": 16,
|
| 30 |
+
"qkv_bias": true,
|
| 31 |
+
"torch_dtype": "float32",
|
| 32 |
+
"transformers_version": "4.45.2"
|
| 33 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a20104ebd7177ed7afca77493358e896bc3f93a3a7dd76098997f95c195d626c
|
| 3 |
+
size 343227052
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:86d38147a77d93ba65d84613f146533d5ad8ed94253d1aa1c91c262dd04e3d94
|
| 3 |
+
size 397031674
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"do_rescale": true,
|
| 4 |
+
"do_resize": true,
|
| 5 |
+
"image_mean": [
|
| 6 |
+
0.5,
|
| 7 |
+
0.5,
|
| 8 |
+
0.5
|
| 9 |
+
],
|
| 10 |
+
"image_processor_type": "ViTImageProcessorFast",
|
| 11 |
+
"image_std": [
|
| 12 |
+
0.5,
|
| 13 |
+
0.5,
|
| 14 |
+
0.5
|
| 15 |
+
],
|
| 16 |
+
"resample": 2,
|
| 17 |
+
"rescale_factor": 0.00392156862745098,
|
| 18 |
+
"size": {
|
| 19 |
+
"height": 224,
|
| 20 |
+
"width": 224
|
| 21 |
+
}
|
| 22 |
+
}
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3abd336be5aa19dd934711acf6c792e4d4905d664d9c20eb143ff4803f81029d
|
| 3 |
+
size 13990
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:35c8ac80e2e3004740b7e835c921ab149674c56cfcd7aa1d0520db45bce1df81
|
| 3 |
+
size 1064
|
trainer_state.json
ADDED
|
@@ -0,0 +1,885 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.728,
|
| 3 |
+
"best_model_checkpoint": "VT_15/checkpoint-7279",
|
| 4 |
+
"epoch": 29.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 7279,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.398406374501992,
|
| 13 |
+
"grad_norm": 11.727704048156738,
|
| 14 |
+
"learning_rate": 9.867197875166003e-05,
|
| 15 |
+
"loss": 1.0157,
|
| 16 |
+
"step": 100
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.796812749003984,
|
| 20 |
+
"grad_norm": 11.50269603729248,
|
| 21 |
+
"learning_rate": 9.734395750332006e-05,
|
| 22 |
+
"loss": 0.8938,
|
| 23 |
+
"step": 200
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 1.0,
|
| 27 |
+
"eval_accuracy": 0.548,
|
| 28 |
+
"eval_f1": 0.4882155949880584,
|
| 29 |
+
"eval_loss": 0.9423586130142212,
|
| 30 |
+
"eval_precision": 0.5733499456903712,
|
| 31 |
+
"eval_recall": 0.5448468191272983,
|
| 32 |
+
"eval_runtime": 16.5911,
|
| 33 |
+
"eval_samples_per_second": 30.137,
|
| 34 |
+
"eval_steps_per_second": 3.797,
|
| 35 |
+
"step": 251
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 1.1952191235059761,
|
| 39 |
+
"grad_norm": 10.47740650177002,
|
| 40 |
+
"learning_rate": 9.601593625498009e-05,
|
| 41 |
+
"loss": 0.8485,
|
| 42 |
+
"step": 300
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 1.593625498007968,
|
| 46 |
+
"grad_norm": 12.222431182861328,
|
| 47 |
+
"learning_rate": 9.468791500664011e-05,
|
| 48 |
+
"loss": 0.7978,
|
| 49 |
+
"step": 400
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 1.9920318725099602,
|
| 53 |
+
"grad_norm": 13.02956771850586,
|
| 54 |
+
"learning_rate": 9.335989375830013e-05,
|
| 55 |
+
"loss": 0.8093,
|
| 56 |
+
"step": 500
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 2.0,
|
| 60 |
+
"eval_accuracy": 0.61,
|
| 61 |
+
"eval_f1": 0.584707628587346,
|
| 62 |
+
"eval_loss": 0.8303987979888916,
|
| 63 |
+
"eval_precision": 0.6022599150943656,
|
| 64 |
+
"eval_recall": 0.6097290164964017,
|
| 65 |
+
"eval_runtime": 15.8756,
|
| 66 |
+
"eval_samples_per_second": 31.495,
|
| 67 |
+
"eval_steps_per_second": 3.968,
|
| 68 |
+
"step": 502
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 2.3904382470119523,
|
| 72 |
+
"grad_norm": 11.920299530029297,
|
| 73 |
+
"learning_rate": 9.203187250996016e-05,
|
| 74 |
+
"loss": 0.7275,
|
| 75 |
+
"step": 600
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 2.7888446215139444,
|
| 79 |
+
"grad_norm": 12.057920455932617,
|
| 80 |
+
"learning_rate": 9.070385126162018e-05,
|
| 81 |
+
"loss": 0.7444,
|
| 82 |
+
"step": 700
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 3.0,
|
| 86 |
+
"eval_accuracy": 0.578,
|
| 87 |
+
"eval_f1": 0.5180026990553307,
|
| 88 |
+
"eval_loss": 0.8966869115829468,
|
| 89 |
+
"eval_precision": 0.6246163183733936,
|
| 90 |
+
"eval_recall": 0.5751176873923239,
|
| 91 |
+
"eval_runtime": 16.5617,
|
| 92 |
+
"eval_samples_per_second": 30.19,
|
| 93 |
+
"eval_steps_per_second": 3.804,
|
| 94 |
+
"step": 753
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 3.187250996015936,
|
| 98 |
+
"grad_norm": 12.730193138122559,
|
| 99 |
+
"learning_rate": 8.937583001328021e-05,
|
| 100 |
+
"loss": 0.7091,
|
| 101 |
+
"step": 800
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 3.585657370517928,
|
| 105 |
+
"grad_norm": 11.721458435058594,
|
| 106 |
+
"learning_rate": 8.804780876494024e-05,
|
| 107 |
+
"loss": 0.6502,
|
| 108 |
+
"step": 900
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 3.9840637450199203,
|
| 112 |
+
"grad_norm": 11.902241706848145,
|
| 113 |
+
"learning_rate": 8.671978751660027e-05,
|
| 114 |
+
"loss": 0.6391,
|
| 115 |
+
"step": 1000
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 4.0,
|
| 119 |
+
"eval_accuracy": 0.624,
|
| 120 |
+
"eval_f1": 0.6176260916415671,
|
| 121 |
+
"eval_loss": 0.8131240010261536,
|
| 122 |
+
"eval_precision": 0.6213088498802785,
|
| 123 |
+
"eval_recall": 0.6225291409540601,
|
| 124 |
+
"eval_runtime": 16.9186,
|
| 125 |
+
"eval_samples_per_second": 29.553,
|
| 126 |
+
"eval_steps_per_second": 3.724,
|
| 127 |
+
"step": 1004
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 4.382470119521912,
|
| 131 |
+
"grad_norm": 11.794739723205566,
|
| 132 |
+
"learning_rate": 8.539176626826029e-05,
|
| 133 |
+
"loss": 0.5683,
|
| 134 |
+
"step": 1100
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 4.780876494023905,
|
| 138 |
+
"grad_norm": 11.683808326721191,
|
| 139 |
+
"learning_rate": 8.406374501992032e-05,
|
| 140 |
+
"loss": 0.5691,
|
| 141 |
+
"step": 1200
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 5.0,
|
| 145 |
+
"eval_accuracy": 0.622,
|
| 146 |
+
"eval_f1": 0.5716700610476999,
|
| 147 |
+
"eval_loss": 0.8928351998329163,
|
| 148 |
+
"eval_precision": 0.6410496659546192,
|
| 149 |
+
"eval_recall": 0.6199959090306707,
|
| 150 |
+
"eval_runtime": 16.6354,
|
| 151 |
+
"eval_samples_per_second": 30.056,
|
| 152 |
+
"eval_steps_per_second": 3.787,
|
| 153 |
+
"step": 1255
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 5.179282868525896,
|
| 157 |
+
"grad_norm": 13.705164909362793,
|
| 158 |
+
"learning_rate": 8.273572377158035e-05,
|
| 159 |
+
"loss": 0.5855,
|
| 160 |
+
"step": 1300
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 5.577689243027889,
|
| 164 |
+
"grad_norm": 11.817400932312012,
|
| 165 |
+
"learning_rate": 8.140770252324038e-05,
|
| 166 |
+
"loss": 0.513,
|
| 167 |
+
"step": 1400
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 5.9760956175298805,
|
| 171 |
+
"grad_norm": 11.755231857299805,
|
| 172 |
+
"learning_rate": 8.00796812749004e-05,
|
| 173 |
+
"loss": 0.5009,
|
| 174 |
+
"step": 1500
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 6.0,
|
| 178 |
+
"eval_accuracy": 0.656,
|
| 179 |
+
"eval_f1": 0.6379970132206839,
|
| 180 |
+
"eval_loss": 0.8215978741645813,
|
| 181 |
+
"eval_precision": 0.6485419139105967,
|
| 182 |
+
"eval_recall": 0.6545787165880164,
|
| 183 |
+
"eval_runtime": 17.7464,
|
| 184 |
+
"eval_samples_per_second": 28.175,
|
| 185 |
+
"eval_steps_per_second": 3.55,
|
| 186 |
+
"step": 1506
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 6.374501992031872,
|
| 190 |
+
"grad_norm": 10.913691520690918,
|
| 191 |
+
"learning_rate": 7.875166002656043e-05,
|
| 192 |
+
"loss": 0.4488,
|
| 193 |
+
"step": 1600
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 6.772908366533865,
|
| 197 |
+
"grad_norm": 10.615025520324707,
|
| 198 |
+
"learning_rate": 7.742363877822046e-05,
|
| 199 |
+
"loss": 0.4855,
|
| 200 |
+
"step": 1700
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 7.0,
|
| 204 |
+
"eval_accuracy": 0.654,
|
| 205 |
+
"eval_f1": 0.6439337408377656,
|
| 206 |
+
"eval_loss": 0.8311923742294312,
|
| 207 |
+
"eval_precision": 0.6495670995670996,
|
| 208 |
+
"eval_recall": 0.6520354012895196,
|
| 209 |
+
"eval_runtime": 17.519,
|
| 210 |
+
"eval_samples_per_second": 28.54,
|
| 211 |
+
"eval_steps_per_second": 3.596,
|
| 212 |
+
"step": 1757
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 7.171314741035856,
|
| 216 |
+
"grad_norm": 12.026023864746094,
|
| 217 |
+
"learning_rate": 7.609561752988048e-05,
|
| 218 |
+
"loss": 0.4177,
|
| 219 |
+
"step": 1800
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 7.569721115537849,
|
| 223 |
+
"grad_norm": 10.010376930236816,
|
| 224 |
+
"learning_rate": 7.476759628154051e-05,
|
| 225 |
+
"loss": 0.409,
|
| 226 |
+
"step": 1900
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 7.968127490039841,
|
| 230 |
+
"grad_norm": 12.596341133117676,
|
| 231 |
+
"learning_rate": 7.343957503320054e-05,
|
| 232 |
+
"loss": 0.39,
|
| 233 |
+
"step": 2000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 8.0,
|
| 237 |
+
"eval_accuracy": 0.682,
|
| 238 |
+
"eval_f1": 0.6538606492353024,
|
| 239 |
+
"eval_loss": 0.9214051365852356,
|
| 240 |
+
"eval_precision": 0.7101278814728985,
|
| 241 |
+
"eval_recall": 0.6804800262743945,
|
| 242 |
+
"eval_runtime": 17.2651,
|
| 243 |
+
"eval_samples_per_second": 28.96,
|
| 244 |
+
"eval_steps_per_second": 3.649,
|
| 245 |
+
"step": 2008
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 8.366533864541832,
|
| 249 |
+
"grad_norm": 14.169309616088867,
|
| 250 |
+
"learning_rate": 7.211155378486057e-05,
|
| 251 |
+
"loss": 0.3646,
|
| 252 |
+
"step": 2100
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 8.764940239043824,
|
| 256 |
+
"grad_norm": 12.020890235900879,
|
| 257 |
+
"learning_rate": 7.07835325365206e-05,
|
| 258 |
+
"loss": 0.3708,
|
| 259 |
+
"step": 2200
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 9.0,
|
| 263 |
+
"eval_accuracy": 0.598,
|
| 264 |
+
"eval_f1": 0.5725541685903895,
|
| 265 |
+
"eval_loss": 1.0422428846359253,
|
| 266 |
+
"eval_precision": 0.6090800979488745,
|
| 267 |
+
"eval_recall": 0.5948744763847355,
|
| 268 |
+
"eval_runtime": 17.2984,
|
| 269 |
+
"eval_samples_per_second": 28.904,
|
| 270 |
+
"eval_steps_per_second": 3.642,
|
| 271 |
+
"step": 2259
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 9.163346613545817,
|
| 275 |
+
"grad_norm": 10.583425521850586,
|
| 276 |
+
"learning_rate": 6.945551128818062e-05,
|
| 277 |
+
"loss": 0.3533,
|
| 278 |
+
"step": 2300
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 9.56175298804781,
|
| 282 |
+
"grad_norm": 13.7178316116333,
|
| 283 |
+
"learning_rate": 6.812749003984064e-05,
|
| 284 |
+
"loss": 0.2984,
|
| 285 |
+
"step": 2400
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 9.9601593625498,
|
| 289 |
+
"grad_norm": 10.42063045501709,
|
| 290 |
+
"learning_rate": 6.679946879150066e-05,
|
| 291 |
+
"loss": 0.3328,
|
| 292 |
+
"step": 2500
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 10.0,
|
| 296 |
+
"eval_accuracy": 0.718,
|
| 297 |
+
"eval_f1": 0.715429962270956,
|
| 298 |
+
"eval_loss": 0.7483692765235901,
|
| 299 |
+
"eval_precision": 0.7195966559320596,
|
| 300 |
+
"eval_recall": 0.7178075285359515,
|
| 301 |
+
"eval_runtime": 17.492,
|
| 302 |
+
"eval_samples_per_second": 28.584,
|
| 303 |
+
"eval_steps_per_second": 3.602,
|
| 304 |
+
"step": 2510
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 10.358565737051793,
|
| 308 |
+
"grad_norm": 10.752534866333008,
|
| 309 |
+
"learning_rate": 6.547144754316069e-05,
|
| 310 |
+
"loss": 0.2561,
|
| 311 |
+
"step": 2600
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 10.756972111553784,
|
| 315 |
+
"grad_norm": 10.988365173339844,
|
| 316 |
+
"learning_rate": 6.414342629482072e-05,
|
| 317 |
+
"loss": 0.3092,
|
| 318 |
+
"step": 2700
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 11.0,
|
| 322 |
+
"eval_accuracy": 0.72,
|
| 323 |
+
"eval_f1": 0.7160176967190494,
|
| 324 |
+
"eval_loss": 0.8250208497047424,
|
| 325 |
+
"eval_precision": 0.7184132303947973,
|
| 326 |
+
"eval_recall": 0.7186398391269527,
|
| 327 |
+
"eval_runtime": 17.2463,
|
| 328 |
+
"eval_samples_per_second": 28.992,
|
| 329 |
+
"eval_steps_per_second": 3.653,
|
| 330 |
+
"step": 2761
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 11.155378486055778,
|
| 334 |
+
"grad_norm": 10.967025756835938,
|
| 335 |
+
"learning_rate": 6.281540504648075e-05,
|
| 336 |
+
"loss": 0.28,
|
| 337 |
+
"step": 2800
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 11.55378486055777,
|
| 341 |
+
"grad_norm": 11.933313369750977,
|
| 342 |
+
"learning_rate": 6.148738379814077e-05,
|
| 343 |
+
"loss": 0.2747,
|
| 344 |
+
"step": 2900
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 11.952191235059761,
|
| 348 |
+
"grad_norm": 12.90857219696045,
|
| 349 |
+
"learning_rate": 6.01593625498008e-05,
|
| 350 |
+
"loss": 0.281,
|
| 351 |
+
"step": 3000
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 12.0,
|
| 355 |
+
"eval_accuracy": 0.694,
|
| 356 |
+
"eval_f1": 0.6781251589992235,
|
| 357 |
+
"eval_loss": 0.9806899428367615,
|
| 358 |
+
"eval_precision": 0.6976797604396068,
|
| 359 |
+
"eval_recall": 0.6921398650556313,
|
| 360 |
+
"eval_runtime": 17.2981,
|
| 361 |
+
"eval_samples_per_second": 28.905,
|
| 362 |
+
"eval_steps_per_second": 3.642,
|
| 363 |
+
"step": 3012
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 12.350597609561753,
|
| 367 |
+
"grad_norm": 11.363082885742188,
|
| 368 |
+
"learning_rate": 5.883134130146083e-05,
|
| 369 |
+
"loss": 0.2596,
|
| 370 |
+
"step": 3100
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 12.749003984063744,
|
| 374 |
+
"grad_norm": 12.736093521118164,
|
| 375 |
+
"learning_rate": 5.7503320053120855e-05,
|
| 376 |
+
"loss": 0.2162,
|
| 377 |
+
"step": 3200
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 13.0,
|
| 381 |
+
"eval_accuracy": 0.708,
|
| 382 |
+
"eval_f1": 0.7021505447248022,
|
| 383 |
+
"eval_loss": 0.9850034117698669,
|
| 384 |
+
"eval_precision": 0.7101112865680695,
|
| 385 |
+
"eval_recall": 0.7063781669000248,
|
| 386 |
+
"eval_runtime": 17.6848,
|
| 387 |
+
"eval_samples_per_second": 28.273,
|
| 388 |
+
"eval_steps_per_second": 3.562,
|
| 389 |
+
"step": 3263
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 13.147410358565738,
|
| 393 |
+
"grad_norm": 11.957535743713379,
|
| 394 |
+
"learning_rate": 5.6175298804780876e-05,
|
| 395 |
+
"loss": 0.2271,
|
| 396 |
+
"step": 3300
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 13.54581673306773,
|
| 400 |
+
"grad_norm": 13.91019058227539,
|
| 401 |
+
"learning_rate": 5.48472775564409e-05,
|
| 402 |
+
"loss": 0.2284,
|
| 403 |
+
"step": 3400
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 13.944223107569721,
|
| 407 |
+
"grad_norm": 10.606439590454102,
|
| 408 |
+
"learning_rate": 5.351925630810093e-05,
|
| 409 |
+
"loss": 0.2352,
|
| 410 |
+
"step": 3500
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 14.0,
|
| 414 |
+
"eval_accuracy": 0.698,
|
| 415 |
+
"eval_f1": 0.6875106838790609,
|
| 416 |
+
"eval_loss": 0.9122900366783142,
|
| 417 |
+
"eval_precision": 0.6937544840437923,
|
| 418 |
+
"eval_recall": 0.6964621758194903,
|
| 419 |
+
"eval_runtime": 17.3918,
|
| 420 |
+
"eval_samples_per_second": 28.749,
|
| 421 |
+
"eval_steps_per_second": 3.622,
|
| 422 |
+
"step": 3514
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 14.342629482071713,
|
| 426 |
+
"grad_norm": 10.729408264160156,
|
| 427 |
+
"learning_rate": 5.219123505976096e-05,
|
| 428 |
+
"loss": 0.2339,
|
| 429 |
+
"step": 3600
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 14.741035856573705,
|
| 433 |
+
"grad_norm": 10.717667579650879,
|
| 434 |
+
"learning_rate": 5.0863213811420985e-05,
|
| 435 |
+
"loss": 0.1947,
|
| 436 |
+
"step": 3700
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 15.0,
|
| 440 |
+
"eval_accuracy": 0.694,
|
| 441 |
+
"eval_f1": 0.6847697638967624,
|
| 442 |
+
"eval_loss": 1.0269464254379272,
|
| 443 |
+
"eval_precision": 0.6984319398216817,
|
| 444 |
+
"eval_recall": 0.6918425495381815,
|
| 445 |
+
"eval_runtime": 17.3612,
|
| 446 |
+
"eval_samples_per_second": 28.8,
|
| 447 |
+
"eval_steps_per_second": 3.629,
|
| 448 |
+
"step": 3765
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 15.139442231075698,
|
| 452 |
+
"grad_norm": 15.637863159179688,
|
| 453 |
+
"learning_rate": 4.953519256308101e-05,
|
| 454 |
+
"loss": 0.2033,
|
| 455 |
+
"step": 3800
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 15.53784860557769,
|
| 459 |
+
"grad_norm": 11.52315902709961,
|
| 460 |
+
"learning_rate": 4.820717131474104e-05,
|
| 461 |
+
"loss": 0.2073,
|
| 462 |
+
"step": 3900
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 15.936254980079681,
|
| 466 |
+
"grad_norm": 10.585031509399414,
|
| 467 |
+
"learning_rate": 4.687915006640107e-05,
|
| 468 |
+
"loss": 0.1902,
|
| 469 |
+
"step": 4000
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 16.0,
|
| 473 |
+
"eval_accuracy": 0.702,
|
| 474 |
+
"eval_f1": 0.6936988146223305,
|
| 475 |
+
"eval_loss": 1.0092582702636719,
|
| 476 |
+
"eval_precision": 0.704450992084371,
|
| 477 |
+
"eval_recall": 0.6998746780522377,
|
| 478 |
+
"eval_runtime": 17.2437,
|
| 479 |
+
"eval_samples_per_second": 28.996,
|
| 480 |
+
"eval_steps_per_second": 3.654,
|
| 481 |
+
"step": 4016
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 16.334661354581673,
|
| 485 |
+
"grad_norm": 13.553791046142578,
|
| 486 |
+
"learning_rate": 4.555112881806109e-05,
|
| 487 |
+
"loss": 0.1961,
|
| 488 |
+
"step": 4100
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 16.733067729083665,
|
| 492 |
+
"grad_norm": 10.33850383758545,
|
| 493 |
+
"learning_rate": 4.4223107569721116e-05,
|
| 494 |
+
"loss": 0.1912,
|
| 495 |
+
"step": 4200
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 17.0,
|
| 499 |
+
"eval_accuracy": 0.712,
|
| 500 |
+
"eval_f1": 0.7037256290589013,
|
| 501 |
+
"eval_loss": 0.9451501369476318,
|
| 502 |
+
"eval_precision": 0.7113280708081392,
|
| 503 |
+
"eval_recall": 0.710350440499444,
|
| 504 |
+
"eval_runtime": 17.5777,
|
| 505 |
+
"eval_samples_per_second": 28.445,
|
| 506 |
+
"eval_steps_per_second": 3.584,
|
| 507 |
+
"step": 4267
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 17.131474103585656,
|
| 511 |
+
"grad_norm": 10.550813674926758,
|
| 512 |
+
"learning_rate": 4.289508632138114e-05,
|
| 513 |
+
"loss": 0.1724,
|
| 514 |
+
"step": 4300
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 17.529880478087648,
|
| 518 |
+
"grad_norm": 11.269770622253418,
|
| 519 |
+
"learning_rate": 4.156706507304117e-05,
|
| 520 |
+
"loss": 0.166,
|
| 521 |
+
"step": 4400
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 17.92828685258964,
|
| 525 |
+
"grad_norm": 11.22702693939209,
|
| 526 |
+
"learning_rate": 4.02390438247012e-05,
|
| 527 |
+
"loss": 0.1626,
|
| 528 |
+
"step": 4500
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 18.0,
|
| 532 |
+
"eval_accuracy": 0.71,
|
| 533 |
+
"eval_f1": 0.6965991557822268,
|
| 534 |
+
"eval_loss": 1.0229520797729492,
|
| 535 |
+
"eval_precision": 0.7119117791531585,
|
| 536 |
+
"eval_recall": 0.7080606500607883,
|
| 537 |
+
"eval_runtime": 17.3563,
|
| 538 |
+
"eval_samples_per_second": 28.808,
|
| 539 |
+
"eval_steps_per_second": 3.63,
|
| 540 |
+
"step": 4518
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 18.326693227091635,
|
| 544 |
+
"grad_norm": 11.750710487365723,
|
| 545 |
+
"learning_rate": 3.8911022576361225e-05,
|
| 546 |
+
"loss": 0.146,
|
| 547 |
+
"step": 4600
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 18.725099601593627,
|
| 551 |
+
"grad_norm": 10.37628173828125,
|
| 552 |
+
"learning_rate": 3.758300132802125e-05,
|
| 553 |
+
"loss": 0.1524,
|
| 554 |
+
"step": 4700
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 19.0,
|
| 558 |
+
"eval_accuracy": 0.716,
|
| 559 |
+
"eval_f1": 0.7120852228214192,
|
| 560 |
+
"eval_loss": 0.9977978467941284,
|
| 561 |
+
"eval_precision": 0.7210033022533023,
|
| 562 |
+
"eval_recall": 0.7144085668354911,
|
| 563 |
+
"eval_runtime": 17.5249,
|
| 564 |
+
"eval_samples_per_second": 28.531,
|
| 565 |
+
"eval_steps_per_second": 3.595,
|
| 566 |
+
"step": 4769
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 19.12350597609562,
|
| 570 |
+
"grad_norm": 12.99516487121582,
|
| 571 |
+
"learning_rate": 3.625498007968128e-05,
|
| 572 |
+
"loss": 0.1321,
|
| 573 |
+
"step": 4800
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 19.52191235059761,
|
| 577 |
+
"grad_norm": 11.697456359863281,
|
| 578 |
+
"learning_rate": 3.492695883134131e-05,
|
| 579 |
+
"loss": 0.1508,
|
| 580 |
+
"step": 4900
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"epoch": 19.9203187250996,
|
| 584 |
+
"grad_norm": 11.452008247375488,
|
| 585 |
+
"learning_rate": 3.359893758300133e-05,
|
| 586 |
+
"loss": 0.1258,
|
| 587 |
+
"step": 5000
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 20.0,
|
| 591 |
+
"eval_accuracy": 0.71,
|
| 592 |
+
"eval_f1": 0.7074786456937486,
|
| 593 |
+
"eval_loss": 1.050653338432312,
|
| 594 |
+
"eval_precision": 0.7215994120996662,
|
| 595 |
+
"eval_recall": 0.7083547965174904,
|
| 596 |
+
"eval_runtime": 29.4041,
|
| 597 |
+
"eval_samples_per_second": 17.004,
|
| 598 |
+
"eval_steps_per_second": 2.143,
|
| 599 |
+
"step": 5020
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 20.318725099601593,
|
| 603 |
+
"grad_norm": 12.027978897094727,
|
| 604 |
+
"learning_rate": 3.2270916334661356e-05,
|
| 605 |
+
"loss": 0.1387,
|
| 606 |
+
"step": 5100
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 20.717131474103585,
|
| 610 |
+
"grad_norm": 13.427599906921387,
|
| 611 |
+
"learning_rate": 3.094289508632138e-05,
|
| 612 |
+
"loss": 0.1116,
|
| 613 |
+
"step": 5200
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 21.0,
|
| 617 |
+
"eval_accuracy": 0.724,
|
| 618 |
+
"eval_f1": 0.7153409174915838,
|
| 619 |
+
"eval_loss": 1.0689764022827148,
|
| 620 |
+
"eval_precision": 0.7231692880094706,
|
| 621 |
+
"eval_recall": 0.722209642011374,
|
| 622 |
+
"eval_runtime": 17.4789,
|
| 623 |
+
"eval_samples_per_second": 28.606,
|
| 624 |
+
"eval_steps_per_second": 3.604,
|
| 625 |
+
"step": 5271
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 21.115537848605577,
|
| 629 |
+
"grad_norm": 12.410263061523438,
|
| 630 |
+
"learning_rate": 2.961487383798141e-05,
|
| 631 |
+
"loss": 0.1378,
|
| 632 |
+
"step": 5300
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 21.51394422310757,
|
| 636 |
+
"grad_norm": 12.236252784729004,
|
| 637 |
+
"learning_rate": 2.8286852589641438e-05,
|
| 638 |
+
"loss": 0.1181,
|
| 639 |
+
"step": 5400
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 21.91235059760956,
|
| 643 |
+
"grad_norm": 11.362260818481445,
|
| 644 |
+
"learning_rate": 2.6958831341301462e-05,
|
| 645 |
+
"loss": 0.1158,
|
| 646 |
+
"step": 5500
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 22.0,
|
| 650 |
+
"eval_accuracy": 0.702,
|
| 651 |
+
"eval_f1": 0.6967439243675191,
|
| 652 |
+
"eval_loss": 1.1378962993621826,
|
| 653 |
+
"eval_precision": 0.7034445997704206,
|
| 654 |
+
"eval_recall": 0.7004485661440597,
|
| 655 |
+
"eval_runtime": 17.752,
|
| 656 |
+
"eval_samples_per_second": 28.166,
|
| 657 |
+
"eval_steps_per_second": 3.549,
|
| 658 |
+
"step": 5522
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 22.310756972111555,
|
| 662 |
+
"grad_norm": 12.567873001098633,
|
| 663 |
+
"learning_rate": 2.563081009296149e-05,
|
| 664 |
+
"loss": 0.1089,
|
| 665 |
+
"step": 5600
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 22.709163346613547,
|
| 669 |
+
"grad_norm": 10.872307777404785,
|
| 670 |
+
"learning_rate": 2.4302788844621517e-05,
|
| 671 |
+
"loss": 0.1069,
|
| 672 |
+
"step": 5700
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 23.0,
|
| 676 |
+
"eval_accuracy": 0.722,
|
| 677 |
+
"eval_f1": 0.7172758119553166,
|
| 678 |
+
"eval_loss": 1.157360553741455,
|
| 679 |
+
"eval_precision": 0.727205590108816,
|
| 680 |
+
"eval_recall": 0.7205813209797584,
|
| 681 |
+
"eval_runtime": 17.8753,
|
| 682 |
+
"eval_samples_per_second": 27.972,
|
| 683 |
+
"eval_steps_per_second": 3.524,
|
| 684 |
+
"step": 5773
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 23.10756972111554,
|
| 688 |
+
"grad_norm": 10.918773651123047,
|
| 689 |
+
"learning_rate": 2.297476759628154e-05,
|
| 690 |
+
"loss": 0.1112,
|
| 691 |
+
"step": 5800
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 23.50597609561753,
|
| 695 |
+
"grad_norm": 11.303016662597656,
|
| 696 |
+
"learning_rate": 2.1646746347941568e-05,
|
| 697 |
+
"loss": 0.0954,
|
| 698 |
+
"step": 5900
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 23.904382470119522,
|
| 702 |
+
"grad_norm": 10.82700252532959,
|
| 703 |
+
"learning_rate": 2.0318725099601595e-05,
|
| 704 |
+
"loss": 0.1089,
|
| 705 |
+
"step": 6000
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 24.0,
|
| 709 |
+
"eval_accuracy": 0.712,
|
| 710 |
+
"eval_f1": 0.7075426800060708,
|
| 711 |
+
"eval_loss": 1.1160012483596802,
|
| 712 |
+
"eval_precision": 0.7194793034050283,
|
| 713 |
+
"eval_recall": 0.7103697429603638,
|
| 714 |
+
"eval_runtime": 17.7441,
|
| 715 |
+
"eval_samples_per_second": 28.178,
|
| 716 |
+
"eval_steps_per_second": 3.55,
|
| 717 |
+
"step": 6024
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 24.302788844621514,
|
| 721 |
+
"grad_norm": 10.5631742477417,
|
| 722 |
+
"learning_rate": 1.899070385126162e-05,
|
| 723 |
+
"loss": 0.0864,
|
| 724 |
+
"step": 6100
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 24.701195219123505,
|
| 728 |
+
"grad_norm": 10.918201446533203,
|
| 729 |
+
"learning_rate": 1.7662682602921647e-05,
|
| 730 |
+
"loss": 0.0999,
|
| 731 |
+
"step": 6200
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 25.0,
|
| 735 |
+
"eval_accuracy": 0.716,
|
| 736 |
+
"eval_f1": 0.7090935362771184,
|
| 737 |
+
"eval_loss": 1.0727450847625732,
|
| 738 |
+
"eval_precision": 0.7106284520077623,
|
| 739 |
+
"eval_recall": 0.7145099767794276,
|
| 740 |
+
"eval_runtime": 17.8845,
|
| 741 |
+
"eval_samples_per_second": 27.957,
|
| 742 |
+
"eval_steps_per_second": 3.523,
|
| 743 |
+
"step": 6275
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 25.099601593625497,
|
| 747 |
+
"grad_norm": 11.402228355407715,
|
| 748 |
+
"learning_rate": 1.6334661354581674e-05,
|
| 749 |
+
"loss": 0.1042,
|
| 750 |
+
"step": 6300
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 25.49800796812749,
|
| 754 |
+
"grad_norm": 11.740915298461914,
|
| 755 |
+
"learning_rate": 1.5006640106241702e-05,
|
| 756 |
+
"loss": 0.089,
|
| 757 |
+
"step": 6400
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 25.89641434262948,
|
| 761 |
+
"grad_norm": 11.218791961669922,
|
| 762 |
+
"learning_rate": 1.3678618857901726e-05,
|
| 763 |
+
"loss": 0.0738,
|
| 764 |
+
"step": 6500
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 26.0,
|
| 768 |
+
"eval_accuracy": 0.706,
|
| 769 |
+
"eval_f1": 0.699550751079995,
|
| 770 |
+
"eval_loss": 1.2584666013717651,
|
| 771 |
+
"eval_precision": 0.7133200179296525,
|
| 772 |
+
"eval_recall": 0.704105950343699,
|
| 773 |
+
"eval_runtime": 18.0765,
|
| 774 |
+
"eval_samples_per_second": 27.66,
|
| 775 |
+
"eval_steps_per_second": 3.485,
|
| 776 |
+
"step": 6526
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 26.294820717131476,
|
| 780 |
+
"grad_norm": 9.9302339553833,
|
| 781 |
+
"learning_rate": 1.2350597609561753e-05,
|
| 782 |
+
"loss": 0.0914,
|
| 783 |
+
"step": 6600
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 26.693227091633467,
|
| 787 |
+
"grad_norm": 11.051177024841309,
|
| 788 |
+
"learning_rate": 1.102257636122178e-05,
|
| 789 |
+
"loss": 0.0836,
|
| 790 |
+
"step": 6700
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 27.0,
|
| 794 |
+
"eval_accuracy": 0.718,
|
| 795 |
+
"eval_f1": 0.7104355302219595,
|
| 796 |
+
"eval_loss": 1.1709084510803223,
|
| 797 |
+
"eval_precision": 0.7172922964310544,
|
| 798 |
+
"eval_recall": 0.7163028008735083,
|
| 799 |
+
"eval_runtime": 17.5336,
|
| 800 |
+
"eval_samples_per_second": 28.517,
|
| 801 |
+
"eval_steps_per_second": 3.593,
|
| 802 |
+
"step": 6777
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 27.09163346613546,
|
| 806 |
+
"grad_norm": 10.727697372436523,
|
| 807 |
+
"learning_rate": 9.694555112881806e-06,
|
| 808 |
+
"loss": 0.0986,
|
| 809 |
+
"step": 6800
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 27.49003984063745,
|
| 813 |
+
"grad_norm": 11.817888259887695,
|
| 814 |
+
"learning_rate": 8.366533864541832e-06,
|
| 815 |
+
"loss": 0.07,
|
| 816 |
+
"step": 6900
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 27.888446215139442,
|
| 820 |
+
"grad_norm": 10.357769966125488,
|
| 821 |
+
"learning_rate": 7.03851261620186e-06,
|
| 822 |
+
"loss": 0.0775,
|
| 823 |
+
"step": 7000
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 28.0,
|
| 827 |
+
"eval_accuracy": 0.722,
|
| 828 |
+
"eval_f1": 0.7145836341124611,
|
| 829 |
+
"eval_loss": 1.2422434091567993,
|
| 830 |
+
"eval_precision": 0.7256025662918439,
|
| 831 |
+
"eval_recall": 0.720296105512437,
|
| 832 |
+
"eval_runtime": 17.5709,
|
| 833 |
+
"eval_samples_per_second": 28.456,
|
| 834 |
+
"eval_steps_per_second": 3.585,
|
| 835 |
+
"step": 7028
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 28.286852589641434,
|
| 839 |
+
"grad_norm": 10.4796142578125,
|
| 840 |
+
"learning_rate": 5.710491367861886e-06,
|
| 841 |
+
"loss": 0.0713,
|
| 842 |
+
"step": 7100
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 28.685258964143426,
|
| 846 |
+
"grad_norm": 9.803996086120605,
|
| 847 |
+
"learning_rate": 4.382470119521913e-06,
|
| 848 |
+
"loss": 0.0752,
|
| 849 |
+
"step": 7200
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 29.0,
|
| 853 |
+
"eval_accuracy": 0.728,
|
| 854 |
+
"eval_f1": 0.7205734767025089,
|
| 855 |
+
"eval_loss": 1.2145317792892456,
|
| 856 |
+
"eval_precision": 0.730059540405073,
|
| 857 |
+
"eval_recall": 0.7263936664880468,
|
| 858 |
+
"eval_runtime": 17.625,
|
| 859 |
+
"eval_samples_per_second": 28.369,
|
| 860 |
+
"eval_steps_per_second": 3.574,
|
| 861 |
+
"step": 7279
|
| 862 |
+
}
|
| 863 |
+
],
|
| 864 |
+
"logging_steps": 100,
|
| 865 |
+
"max_steps": 7530,
|
| 866 |
+
"num_input_tokens_seen": 0,
|
| 867 |
+
"num_train_epochs": 30,
|
| 868 |
+
"save_steps": 500,
|
| 869 |
+
"stateful_callbacks": {
|
| 870 |
+
"TrainerControl": {
|
| 871 |
+
"args": {
|
| 872 |
+
"should_epoch_stop": false,
|
| 873 |
+
"should_evaluate": false,
|
| 874 |
+
"should_log": false,
|
| 875 |
+
"should_save": true,
|
| 876 |
+
"should_training_stop": false
|
| 877 |
+
},
|
| 878 |
+
"attributes": {}
|
| 879 |
+
}
|
| 880 |
+
},
|
| 881 |
+
"total_flos": 8.995893225012062e+18,
|
| 882 |
+
"train_batch_size": 16,
|
| 883 |
+
"trial_name": null,
|
| 884 |
+
"trial_params": null
|
| 885 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3578ac3aa4d3823d4733a342727e9029e0b82da9c9013e117e3a548971b8a425
|
| 3 |
+
size 5240
|