Upload 4 files
Browse files- config.json +46 -0
- preprocessor_config.json +24 -0
- trainer_state.json +967 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,46 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"SiglipForImageClassification"
|
| 4 |
+
],
|
| 5 |
+
"id2label": {
|
| 6 |
+
"0": "ai",
|
| 7 |
+
"1": "hum"
|
| 8 |
+
},
|
| 9 |
+
"initializer_factor": 1.0,
|
| 10 |
+
"label2id": {
|
| 11 |
+
"ai": 0,
|
| 12 |
+
"hum": 1
|
| 13 |
+
},
|
| 14 |
+
"model_type": "siglip",
|
| 15 |
+
"problem_type": "single_label_classification",
|
| 16 |
+
"text_config": {
|
| 17 |
+
"attention_dropout": 0.0,
|
| 18 |
+
"hidden_act": "gelu_pytorch_tanh",
|
| 19 |
+
"hidden_size": 768,
|
| 20 |
+
"intermediate_size": 3072,
|
| 21 |
+
"layer_norm_eps": 1e-06,
|
| 22 |
+
"max_position_embeddings": 64,
|
| 23 |
+
"model_type": "siglip_text_model",
|
| 24 |
+
"num_attention_heads": 12,
|
| 25 |
+
"num_hidden_layers": 12,
|
| 26 |
+
"projection_size": 768,
|
| 27 |
+
"torch_dtype": "float32",
|
| 28 |
+
"vocab_size": 256000
|
| 29 |
+
},
|
| 30 |
+
"torch_dtype": "float32",
|
| 31 |
+
"transformers_version": "4.51.0.dev0",
|
| 32 |
+
"vision_config": {
|
| 33 |
+
"attention_dropout": 0.0,
|
| 34 |
+
"hidden_act": "gelu_pytorch_tanh",
|
| 35 |
+
"hidden_size": 768,
|
| 36 |
+
"image_size": 224,
|
| 37 |
+
"intermediate_size": 3072,
|
| 38 |
+
"layer_norm_eps": 1e-06,
|
| 39 |
+
"model_type": "siglip_vision_model",
|
| 40 |
+
"num_attention_heads": 12,
|
| 41 |
+
"num_channels": 3,
|
| 42 |
+
"num_hidden_layers": 12,
|
| 43 |
+
"patch_size": 16,
|
| 44 |
+
"torch_dtype": "float32"
|
| 45 |
+
}
|
| 46 |
+
}
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_convert_rgb": null,
|
| 3 |
+
"do_normalize": true,
|
| 4 |
+
"do_rescale": true,
|
| 5 |
+
"do_resize": true,
|
| 6 |
+
"image_mean": [
|
| 7 |
+
0.5,
|
| 8 |
+
0.5,
|
| 9 |
+
0.5
|
| 10 |
+
],
|
| 11 |
+
"image_processor_type": "SiglipImageProcessor",
|
| 12 |
+
"image_std": [
|
| 13 |
+
0.5,
|
| 14 |
+
0.5,
|
| 15 |
+
0.5
|
| 16 |
+
],
|
| 17 |
+
"processor_class": "SiglipProcessor",
|
| 18 |
+
"resample": 2,
|
| 19 |
+
"rescale_factor": 0.00392156862745098,
|
| 20 |
+
"size": {
|
| 21 |
+
"height": 224,
|
| 22 |
+
"width": 224
|
| 23 |
+
}
|
| 24 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,967 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 66218,
|
| 3 |
+
"best_metric": 0.9523090398381104,
|
| 4 |
+
"best_model_checkpoint": "ai-vs-hum\\checkpoint-66218",
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 66218,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.007550816998399227,
|
| 14 |
+
"grad_norm": 65.657958984375,
|
| 15 |
+
"learning_rate": 7.550816998399228e-08,
|
| 16 |
+
"loss": 0.7738,
|
| 17 |
+
"step": 500
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.015101633996798454,
|
| 21 |
+
"grad_norm": 56.08807373046875,
|
| 22 |
+
"learning_rate": 1.5101633996798456e-07,
|
| 23 |
+
"loss": 0.715,
|
| 24 |
+
"step": 1000
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.02265245099519768,
|
| 28 |
+
"grad_norm": 80.93883514404297,
|
| 29 |
+
"learning_rate": 2.265245099519768e-07,
|
| 30 |
+
"loss": 0.6352,
|
| 31 |
+
"step": 1500
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.03020326799359691,
|
| 35 |
+
"grad_norm": 99.87760925292969,
|
| 36 |
+
"learning_rate": 3.020326799359691e-07,
|
| 37 |
+
"loss": 0.5265,
|
| 38 |
+
"step": 2000
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.03775408499199613,
|
| 42 |
+
"grad_norm": 61.885841369628906,
|
| 43 |
+
"learning_rate": 3.7754084991996136e-07,
|
| 44 |
+
"loss": 0.4273,
|
| 45 |
+
"step": 2500
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.04530490199039536,
|
| 49 |
+
"grad_norm": 704.5555419921875,
|
| 50 |
+
"learning_rate": 4.530490199039536e-07,
|
| 51 |
+
"loss": 0.4747,
|
| 52 |
+
"step": 3000
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.05285571898879459,
|
| 56 |
+
"grad_norm": 0.1631409078836441,
|
| 57 |
+
"learning_rate": 5.285571898879459e-07,
|
| 58 |
+
"loss": 0.5541,
|
| 59 |
+
"step": 3500
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.06040653598719382,
|
| 63 |
+
"grad_norm": 0.1648247241973877,
|
| 64 |
+
"learning_rate": 6.040653598719382e-07,
|
| 65 |
+
"loss": 0.7388,
|
| 66 |
+
"step": 4000
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.06795735298559304,
|
| 70 |
+
"grad_norm": 2444.441650390625,
|
| 71 |
+
"learning_rate": 6.795735298559304e-07,
|
| 72 |
+
"loss": 0.6825,
|
| 73 |
+
"step": 4500
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.07550816998399226,
|
| 77 |
+
"grad_norm": 0.37286192178726196,
|
| 78 |
+
"learning_rate": 7.550816998399227e-07,
|
| 79 |
+
"loss": 0.6872,
|
| 80 |
+
"step": 5000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.08305898698239149,
|
| 84 |
+
"grad_norm": 0.28558024764060974,
|
| 85 |
+
"learning_rate": 8.305898698239149e-07,
|
| 86 |
+
"loss": 0.7039,
|
| 87 |
+
"step": 5500
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.09060980398079072,
|
| 91 |
+
"grad_norm": 182.67251586914062,
|
| 92 |
+
"learning_rate": 9.060980398079072e-07,
|
| 93 |
+
"loss": 0.6045,
|
| 94 |
+
"step": 6000
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.09816062097918994,
|
| 98 |
+
"grad_norm": 0.0031864135526120663,
|
| 99 |
+
"learning_rate": 9.816062097918996e-07,
|
| 100 |
+
"loss": 0.5625,
|
| 101 |
+
"step": 6500
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.10571143797758918,
|
| 105 |
+
"grad_norm": 0.03901828080415726,
|
| 106 |
+
"learning_rate": 1.0571143797758919e-06,
|
| 107 |
+
"loss": 0.5507,
|
| 108 |
+
"step": 7000
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.11326225497598841,
|
| 112 |
+
"grad_norm": 0.004638838116079569,
|
| 113 |
+
"learning_rate": 1.1326225497598842e-06,
|
| 114 |
+
"loss": 0.5191,
|
| 115 |
+
"step": 7500
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.12081307197438763,
|
| 119 |
+
"grad_norm": 0.14078642427921295,
|
| 120 |
+
"learning_rate": 1.2081307197438765e-06,
|
| 121 |
+
"loss": 0.5333,
|
| 122 |
+
"step": 8000
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.12836388897278686,
|
| 126 |
+
"grad_norm": 0.07704892754554749,
|
| 127 |
+
"learning_rate": 1.2836388897278688e-06,
|
| 128 |
+
"loss": 0.5258,
|
| 129 |
+
"step": 8500
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.13591470597118607,
|
| 133 |
+
"grad_norm": 0.03960805386304855,
|
| 134 |
+
"learning_rate": 1.3591470597118609e-06,
|
| 135 |
+
"loss": 0.4775,
|
| 136 |
+
"step": 9000
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.1434655229695853,
|
| 140 |
+
"grad_norm": 0.8385183215141296,
|
| 141 |
+
"learning_rate": 1.4346552296958531e-06,
|
| 142 |
+
"loss": 0.4945,
|
| 143 |
+
"step": 9500
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.15101633996798453,
|
| 147 |
+
"grad_norm": 574.739990234375,
|
| 148 |
+
"learning_rate": 1.5101633996798454e-06,
|
| 149 |
+
"loss": 0.5178,
|
| 150 |
+
"step": 10000
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.15856715696638377,
|
| 154 |
+
"grad_norm": 0.06548255681991577,
|
| 155 |
+
"learning_rate": 1.5856715696638377e-06,
|
| 156 |
+
"loss": 0.4735,
|
| 157 |
+
"step": 10500
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.16611797396478298,
|
| 161 |
+
"grad_norm": 1357.1788330078125,
|
| 162 |
+
"learning_rate": 1.6611797396478298e-06,
|
| 163 |
+
"loss": 0.5533,
|
| 164 |
+
"step": 11000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.17366879096318222,
|
| 168 |
+
"grad_norm": 0.12339621037244797,
|
| 169 |
+
"learning_rate": 1.7366879096318223e-06,
|
| 170 |
+
"loss": 0.4891,
|
| 171 |
+
"step": 11500
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.18121960796158143,
|
| 175 |
+
"grad_norm": 0.3505186438560486,
|
| 176 |
+
"learning_rate": 1.8121960796158144e-06,
|
| 177 |
+
"loss": 0.4935,
|
| 178 |
+
"step": 12000
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.18877042495998067,
|
| 182 |
+
"grad_norm": 1131.322509765625,
|
| 183 |
+
"learning_rate": 1.887704249599807e-06,
|
| 184 |
+
"loss": 0.4447,
|
| 185 |
+
"step": 12500
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.19632124195837988,
|
| 189 |
+
"grad_norm": 0.030383553355932236,
|
| 190 |
+
"learning_rate": 1.963212419583799e-06,
|
| 191 |
+
"loss": 0.5052,
|
| 192 |
+
"step": 13000
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.20387205895677912,
|
| 196 |
+
"grad_norm": 0.06966419517993927,
|
| 197 |
+
"learning_rate": 2.0387205895677913e-06,
|
| 198 |
+
"loss": 0.4477,
|
| 199 |
+
"step": 13500
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.21142287595517836,
|
| 203 |
+
"grad_norm": 0.14627192914485931,
|
| 204 |
+
"learning_rate": 2.1142287595517838e-06,
|
| 205 |
+
"loss": 0.4535,
|
| 206 |
+
"step": 14000
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.21897369295357758,
|
| 210 |
+
"grad_norm": 0.013991514220833778,
|
| 211 |
+
"learning_rate": 2.189736929535776e-06,
|
| 212 |
+
"loss": 0.414,
|
| 213 |
+
"step": 14500
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.22652450995197682,
|
| 217 |
+
"grad_norm": 278.3962707519531,
|
| 218 |
+
"learning_rate": 2.2652450995197684e-06,
|
| 219 |
+
"loss": 0.3346,
|
| 220 |
+
"step": 15000
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.23407532695037603,
|
| 224 |
+
"grad_norm": 0.012267685495316982,
|
| 225 |
+
"learning_rate": 2.3407532695037605e-06,
|
| 226 |
+
"loss": 0.426,
|
| 227 |
+
"step": 15500
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.24162614394877527,
|
| 231 |
+
"grad_norm": 0.0350683256983757,
|
| 232 |
+
"learning_rate": 2.416261439487753e-06,
|
| 233 |
+
"loss": 0.3849,
|
| 234 |
+
"step": 16000
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.24917696094717448,
|
| 238 |
+
"grad_norm": 0.01037506852298975,
|
| 239 |
+
"learning_rate": 2.491769609471745e-06,
|
| 240 |
+
"loss": 0.3726,
|
| 241 |
+
"step": 16500
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.2567277779455737,
|
| 245 |
+
"grad_norm": 139.85562133789062,
|
| 246 |
+
"learning_rate": 2.5672777794557375e-06,
|
| 247 |
+
"loss": 0.4721,
|
| 248 |
+
"step": 17000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.26427859494397293,
|
| 252 |
+
"grad_norm": 0.11431169509887695,
|
| 253 |
+
"learning_rate": 2.6427859494397296e-06,
|
| 254 |
+
"loss": 0.4746,
|
| 255 |
+
"step": 17500
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.27182941194237215,
|
| 259 |
+
"grad_norm": 0.024222318083047867,
|
| 260 |
+
"learning_rate": 2.7182941194237217e-06,
|
| 261 |
+
"loss": 0.3612,
|
| 262 |
+
"step": 18000
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.2793802289407714,
|
| 266 |
+
"grad_norm": 0.0600462444126606,
|
| 267 |
+
"learning_rate": 2.793802289407714e-06,
|
| 268 |
+
"loss": 0.3716,
|
| 269 |
+
"step": 18500
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.2869310459391706,
|
| 273 |
+
"grad_norm": 0.0730457752943039,
|
| 274 |
+
"learning_rate": 2.8693104593917063e-06,
|
| 275 |
+
"loss": 0.4519,
|
| 276 |
+
"step": 19000
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.29448186293756984,
|
| 280 |
+
"grad_norm": 0.1063174456357956,
|
| 281 |
+
"learning_rate": 2.944818629375699e-06,
|
| 282 |
+
"loss": 0.3865,
|
| 283 |
+
"step": 19500
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.30203267993596905,
|
| 287 |
+
"grad_norm": 0.009060272946953773,
|
| 288 |
+
"learning_rate": 3.020326799359691e-06,
|
| 289 |
+
"loss": 0.3653,
|
| 290 |
+
"step": 20000
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.3095834969343683,
|
| 294 |
+
"grad_norm": 0.0033603431656956673,
|
| 295 |
+
"learning_rate": 3.0958349693436834e-06,
|
| 296 |
+
"loss": 0.4347,
|
| 297 |
+
"step": 20500
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.31713431393276753,
|
| 301 |
+
"grad_norm": 0.011356896720826626,
|
| 302 |
+
"learning_rate": 3.1713431393276755e-06,
|
| 303 |
+
"loss": 0.3759,
|
| 304 |
+
"step": 21000
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.32468513093116674,
|
| 308 |
+
"grad_norm": 0.2442641258239746,
|
| 309 |
+
"learning_rate": 3.2468513093116675e-06,
|
| 310 |
+
"loss": 0.3776,
|
| 311 |
+
"step": 21500
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.33223594792956596,
|
| 315 |
+
"grad_norm": 0.029270794242620468,
|
| 316 |
+
"learning_rate": 3.3223594792956596e-06,
|
| 317 |
+
"loss": 0.3688,
|
| 318 |
+
"step": 22000
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.3397867649279652,
|
| 322 |
+
"grad_norm": 0.28456059098243713,
|
| 323 |
+
"learning_rate": 3.3978676492796526e-06,
|
| 324 |
+
"loss": 0.3656,
|
| 325 |
+
"step": 22500
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.34733758192636444,
|
| 329 |
+
"grad_norm": 0.07442634552717209,
|
| 330 |
+
"learning_rate": 3.4733758192636446e-06,
|
| 331 |
+
"loss": 0.4154,
|
| 332 |
+
"step": 23000
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.35488839892476365,
|
| 336 |
+
"grad_norm": 0.029278529807925224,
|
| 337 |
+
"learning_rate": 3.5488839892476367e-06,
|
| 338 |
+
"loss": 0.4295,
|
| 339 |
+
"step": 23500
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.36243921592316286,
|
| 343 |
+
"grad_norm": 128.08538818359375,
|
| 344 |
+
"learning_rate": 3.624392159231629e-06,
|
| 345 |
+
"loss": 0.376,
|
| 346 |
+
"step": 24000
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.36999003292156213,
|
| 350 |
+
"grad_norm": 0.05300796404480934,
|
| 351 |
+
"learning_rate": 3.6999003292156217e-06,
|
| 352 |
+
"loss": 0.3514,
|
| 353 |
+
"step": 24500
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.37754084991996134,
|
| 357 |
+
"grad_norm": 0.005980394314974546,
|
| 358 |
+
"learning_rate": 3.775408499199614e-06,
|
| 359 |
+
"loss": 0.4061,
|
| 360 |
+
"step": 25000
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.38509166691836055,
|
| 364 |
+
"grad_norm": 121.62598419189453,
|
| 365 |
+
"learning_rate": 3.8509166691836055e-06,
|
| 366 |
+
"loss": 0.3857,
|
| 367 |
+
"step": 25500
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.39264248391675977,
|
| 371 |
+
"grad_norm": 0.09926415979862213,
|
| 372 |
+
"learning_rate": 3.926424839167598e-06,
|
| 373 |
+
"loss": 0.3912,
|
| 374 |
+
"step": 26000
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.40019330091515903,
|
| 378 |
+
"grad_norm": 0.11781750619411469,
|
| 379 |
+
"learning_rate": 4.0019330091515905e-06,
|
| 380 |
+
"loss": 0.4562,
|
| 381 |
+
"step": 26500
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.40774411791355825,
|
| 385 |
+
"grad_norm": 0.034933462738990784,
|
| 386 |
+
"learning_rate": 4.0774411791355826e-06,
|
| 387 |
+
"loss": 0.3881,
|
| 388 |
+
"step": 27000
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.41529493491195746,
|
| 392 |
+
"grad_norm": 0.07762602716684341,
|
| 393 |
+
"learning_rate": 4.152949349119575e-06,
|
| 394 |
+
"loss": 0.4326,
|
| 395 |
+
"step": 27500
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.4228457519103567,
|
| 399 |
+
"grad_norm": 0.03593170642852783,
|
| 400 |
+
"learning_rate": 4.2284575191035676e-06,
|
| 401 |
+
"loss": 0.4169,
|
| 402 |
+
"step": 28000
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.43039656890875594,
|
| 406 |
+
"grad_norm": 6.517802715301514,
|
| 407 |
+
"learning_rate": 4.30396568908756e-06,
|
| 408 |
+
"loss": 0.3935,
|
| 409 |
+
"step": 28500
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.43794738590715515,
|
| 413 |
+
"grad_norm": 0.08031677454710007,
|
| 414 |
+
"learning_rate": 4.379473859071552e-06,
|
| 415 |
+
"loss": 0.3949,
|
| 416 |
+
"step": 29000
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.44549820290555436,
|
| 420 |
+
"grad_norm": 0.00936515349894762,
|
| 421 |
+
"learning_rate": 4.454982029055544e-06,
|
| 422 |
+
"loss": 0.4852,
|
| 423 |
+
"step": 29500
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.45304901990395363,
|
| 427 |
+
"grad_norm": 0.01062073465436697,
|
| 428 |
+
"learning_rate": 4.530490199039537e-06,
|
| 429 |
+
"loss": 0.3993,
|
| 430 |
+
"step": 30000
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.46059983690235284,
|
| 434 |
+
"grad_norm": 0.0008572549559175968,
|
| 435 |
+
"learning_rate": 4.605998369023529e-06,
|
| 436 |
+
"loss": 0.4309,
|
| 437 |
+
"step": 30500
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.46815065390075206,
|
| 441 |
+
"grad_norm": 0.042835455387830734,
|
| 442 |
+
"learning_rate": 4.681506539007521e-06,
|
| 443 |
+
"loss": 0.5426,
|
| 444 |
+
"step": 31000
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.47570147089915127,
|
| 448 |
+
"grad_norm": 0.006536947563290596,
|
| 449 |
+
"learning_rate": 4.757014708991513e-06,
|
| 450 |
+
"loss": 0.3503,
|
| 451 |
+
"step": 31500
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.48325228789755054,
|
| 455 |
+
"grad_norm": 0.16728435456752777,
|
| 456 |
+
"learning_rate": 4.832522878975506e-06,
|
| 457 |
+
"loss": 0.4025,
|
| 458 |
+
"step": 32000
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.49080310489594975,
|
| 462 |
+
"grad_norm": 0.24009940028190613,
|
| 463 |
+
"learning_rate": 4.908031048959498e-06,
|
| 464 |
+
"loss": 0.373,
|
| 465 |
+
"step": 32500
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.49835392189434896,
|
| 469 |
+
"grad_norm": 0.03488277643918991,
|
| 470 |
+
"learning_rate": 4.98353921894349e-06,
|
| 471 |
+
"loss": 0.4516,
|
| 472 |
+
"step": 33000
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.5059047388927482,
|
| 476 |
+
"grad_norm": 0.07614383846521378,
|
| 477 |
+
"learning_rate": 4.993439179008058e-06,
|
| 478 |
+
"loss": 0.4391,
|
| 479 |
+
"step": 33500
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.5134555558911474,
|
| 483 |
+
"grad_norm": 0.2857000231742859,
|
| 484 |
+
"learning_rate": 4.98504938234317e-06,
|
| 485 |
+
"loss": 0.4202,
|
| 486 |
+
"step": 34000
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.5210063728895467,
|
| 490 |
+
"grad_norm": 0.005722880829125643,
|
| 491 |
+
"learning_rate": 4.9766595856782815e-06,
|
| 492 |
+
"loss": 0.3904,
|
| 493 |
+
"step": 34500
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.5285571898879459,
|
| 497 |
+
"grad_norm": 0.08243716508150101,
|
| 498 |
+
"learning_rate": 4.968269789013394e-06,
|
| 499 |
+
"loss": 0.355,
|
| 500 |
+
"step": 35000
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.5361080068863451,
|
| 504 |
+
"grad_norm": 764.4724731445312,
|
| 505 |
+
"learning_rate": 4.959879992348506e-06,
|
| 506 |
+
"loss": 0.4357,
|
| 507 |
+
"step": 35500
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.5436588238847443,
|
| 511 |
+
"grad_norm": 0.0467102974653244,
|
| 512 |
+
"learning_rate": 4.9514901956836175e-06,
|
| 513 |
+
"loss": 0.4425,
|
| 514 |
+
"step": 36000
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.5512096408831435,
|
| 518 |
+
"grad_norm": 0.4159618318080902,
|
| 519 |
+
"learning_rate": 4.94310039901873e-06,
|
| 520 |
+
"loss": 0.404,
|
| 521 |
+
"step": 36500
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.5587604578815428,
|
| 525 |
+
"grad_norm": 0.0004953582538291812,
|
| 526 |
+
"learning_rate": 4.934710602353842e-06,
|
| 527 |
+
"loss": 0.3587,
|
| 528 |
+
"step": 37000
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.566311274879942,
|
| 532 |
+
"grad_norm": 0.03575737774372101,
|
| 533 |
+
"learning_rate": 4.926320805688954e-06,
|
| 534 |
+
"loss": 0.3726,
|
| 535 |
+
"step": 37500
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.5738620918783413,
|
| 539 |
+
"grad_norm": 0.02385396882891655,
|
| 540 |
+
"learning_rate": 4.917931009024066e-06,
|
| 541 |
+
"loss": 0.3771,
|
| 542 |
+
"step": 38000
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.5814129088767405,
|
| 546 |
+
"grad_norm": 0.02760937251150608,
|
| 547 |
+
"learning_rate": 4.909541212359178e-06,
|
| 548 |
+
"loss": 0.4108,
|
| 549 |
+
"step": 38500
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.5889637258751397,
|
| 553 |
+
"grad_norm": 0.7694607377052307,
|
| 554 |
+
"learning_rate": 4.901151415694289e-06,
|
| 555 |
+
"loss": 0.3603,
|
| 556 |
+
"step": 39000
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.5965145428735389,
|
| 560 |
+
"grad_norm": 1522.3602294921875,
|
| 561 |
+
"learning_rate": 4.892761619029401e-06,
|
| 562 |
+
"loss": 0.3335,
|
| 563 |
+
"step": 39500
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.6040653598719381,
|
| 567 |
+
"grad_norm": 0.3060073256492615,
|
| 568 |
+
"learning_rate": 4.884371822364514e-06,
|
| 569 |
+
"loss": 0.3266,
|
| 570 |
+
"step": 40000
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.6116161768703374,
|
| 574 |
+
"grad_norm": 0.07775181531906128,
|
| 575 |
+
"learning_rate": 4.875982025699625e-06,
|
| 576 |
+
"loss": 0.3596,
|
| 577 |
+
"step": 40500
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.6191669938687366,
|
| 581 |
+
"grad_norm": 0.05560746416449547,
|
| 582 |
+
"learning_rate": 4.867592229034738e-06,
|
| 583 |
+
"loss": 0.4614,
|
| 584 |
+
"step": 41000
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.6267178108671358,
|
| 588 |
+
"grad_norm": 0.11702022701501846,
|
| 589 |
+
"learning_rate": 4.85920243236985e-06,
|
| 590 |
+
"loss": 0.4036,
|
| 591 |
+
"step": 41500
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.6342686278655351,
|
| 595 |
+
"grad_norm": 0.20153406262397766,
|
| 596 |
+
"learning_rate": 4.850812635704961e-06,
|
| 597 |
+
"loss": 0.3323,
|
| 598 |
+
"step": 42000
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.6418194448639343,
|
| 602 |
+
"grad_norm": 1046.9708251953125,
|
| 603 |
+
"learning_rate": 4.842422839040073e-06,
|
| 604 |
+
"loss": 0.423,
|
| 605 |
+
"step": 42500
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.6493702618623335,
|
| 609 |
+
"grad_norm": 0.005386498291045427,
|
| 610 |
+
"learning_rate": 4.834033042375186e-06,
|
| 611 |
+
"loss": 0.3607,
|
| 612 |
+
"step": 43000
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.6569210788607327,
|
| 616 |
+
"grad_norm": 0.0019247422460466623,
|
| 617 |
+
"learning_rate": 4.825643245710297e-06,
|
| 618 |
+
"loss": 0.3117,
|
| 619 |
+
"step": 43500
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.6644718958591319,
|
| 623 |
+
"grad_norm": 0.03873920440673828,
|
| 624 |
+
"learning_rate": 4.817253449045409e-06,
|
| 625 |
+
"loss": 0.4158,
|
| 626 |
+
"step": 44000
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.6720227128575312,
|
| 630 |
+
"grad_norm": 0.00777060491964221,
|
| 631 |
+
"learning_rate": 4.808863652380522e-06,
|
| 632 |
+
"loss": 0.3424,
|
| 633 |
+
"step": 44500
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.6795735298559304,
|
| 637 |
+
"grad_norm": 0.0689612627029419,
|
| 638 |
+
"learning_rate": 4.800473855715633e-06,
|
| 639 |
+
"loss": 0.3596,
|
| 640 |
+
"step": 45000
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.6871243468543297,
|
| 644 |
+
"grad_norm": 164.3638916015625,
|
| 645 |
+
"learning_rate": 4.792084059050745e-06,
|
| 646 |
+
"loss": 0.3271,
|
| 647 |
+
"step": 45500
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.6946751638527289,
|
| 651 |
+
"grad_norm": 0.9296920895576477,
|
| 652 |
+
"learning_rate": 4.783694262385858e-06,
|
| 653 |
+
"loss": 0.4008,
|
| 654 |
+
"step": 46000
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.7022259808511281,
|
| 658 |
+
"grad_norm": 0.008974584750831127,
|
| 659 |
+
"learning_rate": 4.775304465720969e-06,
|
| 660 |
+
"loss": 0.3232,
|
| 661 |
+
"step": 46500
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.7097767978495273,
|
| 665 |
+
"grad_norm": 126.27271270751953,
|
| 666 |
+
"learning_rate": 4.766914669056081e-06,
|
| 667 |
+
"loss": 0.3716,
|
| 668 |
+
"step": 47000
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.7173276148479265,
|
| 672 |
+
"grad_norm": 0.003123954404145479,
|
| 673 |
+
"learning_rate": 4.758524872391193e-06,
|
| 674 |
+
"loss": 0.3153,
|
| 675 |
+
"step": 47500
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.7248784318463257,
|
| 679 |
+
"grad_norm": 0.008505255915224552,
|
| 680 |
+
"learning_rate": 4.750135075726304e-06,
|
| 681 |
+
"loss": 0.3477,
|
| 682 |
+
"step": 48000
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.732429248844725,
|
| 686 |
+
"grad_norm": 0.07402774691581726,
|
| 687 |
+
"learning_rate": 4.741745279061417e-06,
|
| 688 |
+
"loss": 0.3588,
|
| 689 |
+
"step": 48500
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.7399800658431243,
|
| 693 |
+
"grad_norm": 0.029175467789173126,
|
| 694 |
+
"learning_rate": 4.733355482396529e-06,
|
| 695 |
+
"loss": 0.353,
|
| 696 |
+
"step": 49000
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.7475308828415235,
|
| 700 |
+
"grad_norm": 229.5810089111328,
|
| 701 |
+
"learning_rate": 4.724965685731641e-06,
|
| 702 |
+
"loss": 0.361,
|
| 703 |
+
"step": 49500
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.7550816998399227,
|
| 707 |
+
"grad_norm": 0.14088426530361176,
|
| 708 |
+
"learning_rate": 4.716575889066753e-06,
|
| 709 |
+
"loss": 0.3988,
|
| 710 |
+
"step": 50000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.7626325168383219,
|
| 714 |
+
"grad_norm": 0.5763397216796875,
|
| 715 |
+
"learning_rate": 4.7081860924018655e-06,
|
| 716 |
+
"loss": 0.3646,
|
| 717 |
+
"step": 50500
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.7701833338367211,
|
| 721 |
+
"grad_norm": 4.555429458618164,
|
| 722 |
+
"learning_rate": 4.699796295736977e-06,
|
| 723 |
+
"loss": 0.3308,
|
| 724 |
+
"step": 51000
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.7777341508351203,
|
| 728 |
+
"grad_norm": 0.031139669939875603,
|
| 729 |
+
"learning_rate": 4.691406499072089e-06,
|
| 730 |
+
"loss": 0.3487,
|
| 731 |
+
"step": 51500
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.7852849678335195,
|
| 735 |
+
"grad_norm": 1.3805643320083618,
|
| 736 |
+
"learning_rate": 4.683016702407201e-06,
|
| 737 |
+
"loss": 0.3193,
|
| 738 |
+
"step": 52000
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.7928357848319189,
|
| 742 |
+
"grad_norm": 0.00282275746576488,
|
| 743 |
+
"learning_rate": 4.674626905742312e-06,
|
| 744 |
+
"loss": 0.4172,
|
| 745 |
+
"step": 52500
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.8003866018303181,
|
| 749 |
+
"grad_norm": 0.0008711374830454588,
|
| 750 |
+
"learning_rate": 4.666237109077425e-06,
|
| 751 |
+
"loss": 0.3291,
|
| 752 |
+
"step": 53000
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.8079374188287173,
|
| 756 |
+
"grad_norm": 0.007124012336134911,
|
| 757 |
+
"learning_rate": 4.657847312412537e-06,
|
| 758 |
+
"loss": 0.3363,
|
| 759 |
+
"step": 53500
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.8154882358271165,
|
| 763 |
+
"grad_norm": 0.3629874587059021,
|
| 764 |
+
"learning_rate": 4.649457515747648e-06,
|
| 765 |
+
"loss": 0.3525,
|
| 766 |
+
"step": 54000
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.8230390528255157,
|
| 770 |
+
"grad_norm": 89.42652130126953,
|
| 771 |
+
"learning_rate": 4.641067719082761e-06,
|
| 772 |
+
"loss": 0.3133,
|
| 773 |
+
"step": 54500
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.8305898698239149,
|
| 777 |
+
"grad_norm": 0.0034131056163460016,
|
| 778 |
+
"learning_rate": 4.6326779224178726e-06,
|
| 779 |
+
"loss": 0.3381,
|
| 780 |
+
"step": 55000
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.8381406868223141,
|
| 784 |
+
"grad_norm": 8.344011306762695,
|
| 785 |
+
"learning_rate": 4.624288125752985e-06,
|
| 786 |
+
"loss": 0.2783,
|
| 787 |
+
"step": 55500
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.8456915038207135,
|
| 791 |
+
"grad_norm": 0.015363700687885284,
|
| 792 |
+
"learning_rate": 4.615898329088097e-06,
|
| 793 |
+
"loss": 0.3976,
|
| 794 |
+
"step": 56000
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.8532423208191127,
|
| 798 |
+
"grad_norm": 218.98626708984375,
|
| 799 |
+
"learning_rate": 4.6075085324232085e-06,
|
| 800 |
+
"loss": 0.3916,
|
| 801 |
+
"step": 56500
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.8607931378175119,
|
| 805 |
+
"grad_norm": 0.052222587168216705,
|
| 806 |
+
"learning_rate": 4.59911873575832e-06,
|
| 807 |
+
"loss": 0.3614,
|
| 808 |
+
"step": 57000
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.8683439548159111,
|
| 812 |
+
"grad_norm": 0.042011819779872894,
|
| 813 |
+
"learning_rate": 4.590728939093432e-06,
|
| 814 |
+
"loss": 0.422,
|
| 815 |
+
"step": 57500
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.8758947718143103,
|
| 819 |
+
"grad_norm": 0.2174474447965622,
|
| 820 |
+
"learning_rate": 4.5823391424285445e-06,
|
| 821 |
+
"loss": 0.3086,
|
| 822 |
+
"step": 58000
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.8834455888127095,
|
| 826 |
+
"grad_norm": 0.11423001438379288,
|
| 827 |
+
"learning_rate": 4.573949345763656e-06,
|
| 828 |
+
"loss": 0.3596,
|
| 829 |
+
"step": 58500
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.8909964058111087,
|
| 833 |
+
"grad_norm": 0.1473008543252945,
|
| 834 |
+
"learning_rate": 4.565559549098769e-06,
|
| 835 |
+
"loss": 0.3845,
|
| 836 |
+
"step": 59000
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.8985472228095079,
|
| 840 |
+
"grad_norm": 0.6927057504653931,
|
| 841 |
+
"learning_rate": 4.5571697524338805e-06,
|
| 842 |
+
"loss": 0.3272,
|
| 843 |
+
"step": 59500
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.9060980398079073,
|
| 847 |
+
"grad_norm": 0.052435796707868576,
|
| 848 |
+
"learning_rate": 4.548779955768992e-06,
|
| 849 |
+
"loss": 0.3124,
|
| 850 |
+
"step": 60000
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.9136488568063065,
|
| 854 |
+
"grad_norm": 0.0011638773139566183,
|
| 855 |
+
"learning_rate": 4.540390159104105e-06,
|
| 856 |
+
"loss": 0.3051,
|
| 857 |
+
"step": 60500
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.9211996738047057,
|
| 861 |
+
"grad_norm": 0.002878799568861723,
|
| 862 |
+
"learning_rate": 4.5320003624392165e-06,
|
| 863 |
+
"loss": 0.3196,
|
| 864 |
+
"step": 61000
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.9287504908031049,
|
| 868 |
+
"grad_norm": 0.022700993344187737,
|
| 869 |
+
"learning_rate": 4.523610565774328e-06,
|
| 870 |
+
"loss": 0.3307,
|
| 871 |
+
"step": 61500
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.9363013078015041,
|
| 875 |
+
"grad_norm": 18.183610916137695,
|
| 876 |
+
"learning_rate": 4.51522076910944e-06,
|
| 877 |
+
"loss": 0.3215,
|
| 878 |
+
"step": 62000
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.9438521247999033,
|
| 882 |
+
"grad_norm": 0.006543469615280628,
|
| 883 |
+
"learning_rate": 4.5068309724445524e-06,
|
| 884 |
+
"loss": 0.3398,
|
| 885 |
+
"step": 62500
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.9514029417983025,
|
| 889 |
+
"grad_norm": 0.07925090938806534,
|
| 890 |
+
"learning_rate": 4.498441175779664e-06,
|
| 891 |
+
"loss": 0.2937,
|
| 892 |
+
"step": 63000
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.9589537587967017,
|
| 896 |
+
"grad_norm": 0.1537381261587143,
|
| 897 |
+
"learning_rate": 4.490051379114776e-06,
|
| 898 |
+
"loss": 0.2819,
|
| 899 |
+
"step": 63500
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.9665045757951011,
|
| 903 |
+
"grad_norm": 0.01872635819017887,
|
| 904 |
+
"learning_rate": 4.481661582449888e-06,
|
| 905 |
+
"loss": 0.2671,
|
| 906 |
+
"step": 64000
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.9740553927935003,
|
| 910 |
+
"grad_norm": 0.024023612961173058,
|
| 911 |
+
"learning_rate": 4.473271785785e-06,
|
| 912 |
+
"loss": 0.3206,
|
| 913 |
+
"step": 64500
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.9816062097918995,
|
| 917 |
+
"grad_norm": 0.059889055788517,
|
| 918 |
+
"learning_rate": 4.464881989120113e-06,
|
| 919 |
+
"loss": 0.2976,
|
| 920 |
+
"step": 65000
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.9891570267902987,
|
| 924 |
+
"grad_norm": 0.002819158136844635,
|
| 925 |
+
"learning_rate": 4.456492192455224e-06,
|
| 926 |
+
"loss": 0.3465,
|
| 927 |
+
"step": 65500
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.9967078437886979,
|
| 931 |
+
"grad_norm": 0.02669268473982811,
|
| 932 |
+
"learning_rate": 4.448102395790336e-06,
|
| 933 |
+
"loss": 0.3076,
|
| 934 |
+
"step": 66000
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 1.0,
|
| 938 |
+
"eval_accuracy": 0.9523090398381104,
|
| 939 |
+
"eval_loss": 0.3044677674770355,
|
| 940 |
+
"eval_runtime": 202.6207,
|
| 941 |
+
"eval_samples_per_second": 163.404,
|
| 942 |
+
"eval_steps_per_second": 40.855,
|
| 943 |
+
"step": 66218
|
| 944 |
+
}
|
| 945 |
+
],
|
| 946 |
+
"logging_steps": 500,
|
| 947 |
+
"max_steps": 331090,
|
| 948 |
+
"num_input_tokens_seen": 0,
|
| 949 |
+
"num_train_epochs": 5,
|
| 950 |
+
"save_steps": 500,
|
| 951 |
+
"stateful_callbacks": {
|
| 952 |
+
"TrainerControl": {
|
| 953 |
+
"args": {
|
| 954 |
+
"should_epoch_stop": false,
|
| 955 |
+
"should_evaluate": false,
|
| 956 |
+
"should_log": false,
|
| 957 |
+
"should_save": true,
|
| 958 |
+
"should_training_stop": false
|
| 959 |
+
},
|
| 960 |
+
"attributes": {}
|
| 961 |
+
}
|
| 962 |
+
},
|
| 963 |
+
"total_flos": 1.1092242844591276e+19,
|
| 964 |
+
"train_batch_size": 2,
|
| 965 |
+
"trial_name": null,
|
| 966 |
+
"trial_params": null
|
| 967 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d0372770ae24df1223f85193710713f491d29265ce699d46ca7e1aa409b1c6b3
|
| 3 |
+
size 5368
|