Training in progress, step 32000, checkpoint
Browse files
checkpoint-32000/config.json
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "openai/clip-vit-large-patch14-336",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"CLIPModel"
|
| 5 |
+
],
|
| 6 |
+
"initializer_factor": 1.0,
|
| 7 |
+
"logit_scale_init_value": 2.6592,
|
| 8 |
+
"model_type": "clip",
|
| 9 |
+
"projection_dim": 768,
|
| 10 |
+
"text_config": {
|
| 11 |
+
"dropout": 0.0,
|
| 12 |
+
"hidden_size": 768,
|
| 13 |
+
"intermediate_size": 3072,
|
| 14 |
+
"model_type": "clip_text_model",
|
| 15 |
+
"num_attention_heads": 12,
|
| 16 |
+
"projection_dim": 768
|
| 17 |
+
},
|
| 18 |
+
"torch_dtype": "float32",
|
| 19 |
+
"transformers_version": "4.45.0.dev0",
|
| 20 |
+
"vision_config": {
|
| 21 |
+
"dropout": 0.0,
|
| 22 |
+
"hidden_size": 1024,
|
| 23 |
+
"image_size": 336,
|
| 24 |
+
"intermediate_size": 4096,
|
| 25 |
+
"model_type": "clip_vision_model",
|
| 26 |
+
"num_attention_heads": 16,
|
| 27 |
+
"num_hidden_layers": 24,
|
| 28 |
+
"patch_size": 14,
|
| 29 |
+
"projection_dim": 768
|
| 30 |
+
}
|
| 31 |
+
}
|
checkpoint-32000/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d22922f41cfeb267651369d1772c520c0458b71dcd0205f4fbcf5cf53731046c
|
| 3 |
+
size 1711848436
|
checkpoint-32000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:12691df7de00183829914d5910ce2c692fd9dceb907dbe7865dcdff7e74b3339
|
| 3 |
+
size 3424043887
|
checkpoint-32000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:68da1aa8e94e0a836dfe5a7123ac1cc0047706058338cd06f72f26f51a6baf77
|
| 3 |
+
size 14503
|
checkpoint-32000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:83e588fdd67ecc2875076da86fe6d336c60f9a177dc28eee382162749a391769
|
| 3 |
+
size 623
|
checkpoint-32000/trainer_state.json
ADDED
|
@@ -0,0 +1,993 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.6386470794677734,
|
| 3 |
+
"best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e6l56-l/checkpoint-32000",
|
| 4 |
+
"epoch": 5.895357406042741,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 32000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.09211495946941783,
|
| 13 |
+
"grad_norm": 0.49671515822410583,
|
| 14 |
+
"learning_rate": 4.923237533775486e-06,
|
| 15 |
+
"loss": 0.2159,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.09211495946941783,
|
| 20 |
+
"eval_loss": 1.4818799495697021,
|
| 21 |
+
"eval_runtime": 72.2844,
|
| 22 |
+
"eval_samples_per_second": 16.698,
|
| 23 |
+
"eval_steps_per_second": 2.089,
|
| 24 |
+
"step": 500
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.18422991893883567,
|
| 28 |
+
"grad_norm": 0.0015633106231689453,
|
| 29 |
+
"learning_rate": 4.846475067550971e-06,
|
| 30 |
+
"loss": 0.2648,
|
| 31 |
+
"step": 1000
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.18422991893883567,
|
| 35 |
+
"eval_loss": 1.4225908517837524,
|
| 36 |
+
"eval_runtime": 73.6999,
|
| 37 |
+
"eval_samples_per_second": 16.377,
|
| 38 |
+
"eval_steps_per_second": 2.049,
|
| 39 |
+
"step": 1000
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.2763448784082535,
|
| 43 |
+
"grad_norm": 0.047471947968006134,
|
| 44 |
+
"learning_rate": 4.769712601326456e-06,
|
| 45 |
+
"loss": 0.328,
|
| 46 |
+
"step": 1500
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.2763448784082535,
|
| 50 |
+
"eval_loss": 1.412400245666504,
|
| 51 |
+
"eval_runtime": 74.9915,
|
| 52 |
+
"eval_samples_per_second": 16.095,
|
| 53 |
+
"eval_steps_per_second": 2.014,
|
| 54 |
+
"step": 1500
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.36845983787767134,
|
| 58 |
+
"grad_norm": 3.9284778208248383e-10,
|
| 59 |
+
"learning_rate": 4.692950135101941e-06,
|
| 60 |
+
"loss": 0.1813,
|
| 61 |
+
"step": 2000
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.36845983787767134,
|
| 65 |
+
"eval_loss": 1.2844351530075073,
|
| 66 |
+
"eval_runtime": 74.7649,
|
| 67 |
+
"eval_samples_per_second": 16.144,
|
| 68 |
+
"eval_steps_per_second": 2.02,
|
| 69 |
+
"step": 2000
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.46057479734708917,
|
| 73 |
+
"grad_norm": 514.6055297851562,
|
| 74 |
+
"learning_rate": 4.616187668877426e-06,
|
| 75 |
+
"loss": 0.4836,
|
| 76 |
+
"step": 2500
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 0.46057479734708917,
|
| 80 |
+
"eval_loss": 1.301351547241211,
|
| 81 |
+
"eval_runtime": 74.7994,
|
| 82 |
+
"eval_samples_per_second": 16.136,
|
| 83 |
+
"eval_steps_per_second": 2.019,
|
| 84 |
+
"step": 2500
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.552689756816507,
|
| 88 |
+
"grad_norm": 127.98405456542969,
|
| 89 |
+
"learning_rate": 4.5394252026529115e-06,
|
| 90 |
+
"loss": 0.2803,
|
| 91 |
+
"step": 3000
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.552689756816507,
|
| 95 |
+
"eval_loss": 1.1920161247253418,
|
| 96 |
+
"eval_runtime": 75.5447,
|
| 97 |
+
"eval_samples_per_second": 15.977,
|
| 98 |
+
"eval_steps_per_second": 1.999,
|
| 99 |
+
"step": 3000
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.6448047162859248,
|
| 103 |
+
"grad_norm": 0.048586923629045486,
|
| 104 |
+
"learning_rate": 4.462662736428396e-06,
|
| 105 |
+
"loss": 0.2056,
|
| 106 |
+
"step": 3500
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.6448047162859248,
|
| 110 |
+
"eval_loss": 1.1428992748260498,
|
| 111 |
+
"eval_runtime": 75.7202,
|
| 112 |
+
"eval_samples_per_second": 15.94,
|
| 113 |
+
"eval_steps_per_second": 1.994,
|
| 114 |
+
"step": 3500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.7369196757553427,
|
| 118 |
+
"grad_norm": 228.63064575195312,
|
| 119 |
+
"learning_rate": 4.385900270203882e-06,
|
| 120 |
+
"loss": 0.2174,
|
| 121 |
+
"step": 4000
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.7369196757553427,
|
| 125 |
+
"eval_loss": 1.1957736015319824,
|
| 126 |
+
"eval_runtime": 74.1933,
|
| 127 |
+
"eval_samples_per_second": 16.268,
|
| 128 |
+
"eval_steps_per_second": 2.035,
|
| 129 |
+
"step": 4000
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.8290346352247605,
|
| 133 |
+
"grad_norm": 226.2799530029297,
|
| 134 |
+
"learning_rate": 4.3091378039793666e-06,
|
| 135 |
+
"loss": 0.2236,
|
| 136 |
+
"step": 4500
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.8290346352247605,
|
| 140 |
+
"eval_loss": 1.0961644649505615,
|
| 141 |
+
"eval_runtime": 74.6968,
|
| 142 |
+
"eval_samples_per_second": 16.159,
|
| 143 |
+
"eval_steps_per_second": 2.022,
|
| 144 |
+
"step": 4500
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.9211495946941783,
|
| 148 |
+
"grad_norm": 8.084862640478363e-12,
|
| 149 |
+
"learning_rate": 4.232375337754851e-06,
|
| 150 |
+
"loss": 0.1966,
|
| 151 |
+
"step": 5000
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.9211495946941783,
|
| 155 |
+
"eval_loss": 1.1026136875152588,
|
| 156 |
+
"eval_runtime": 75.566,
|
| 157 |
+
"eval_samples_per_second": 15.973,
|
| 158 |
+
"eval_steps_per_second": 1.998,
|
| 159 |
+
"step": 5000
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 1.013264554163596,
|
| 163 |
+
"grad_norm": 1.592839794284373e-06,
|
| 164 |
+
"learning_rate": 4.155612871530337e-06,
|
| 165 |
+
"loss": 0.2547,
|
| 166 |
+
"step": 5500
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 1.013264554163596,
|
| 170 |
+
"eval_loss": 1.0125067234039307,
|
| 171 |
+
"eval_runtime": 74.4786,
|
| 172 |
+
"eval_samples_per_second": 16.206,
|
| 173 |
+
"eval_steps_per_second": 2.027,
|
| 174 |
+
"step": 5500
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 1.105379513633014,
|
| 178 |
+
"grad_norm": 0.06377054750919342,
|
| 179 |
+
"learning_rate": 4.078850405305822e-06,
|
| 180 |
+
"loss": 0.1057,
|
| 181 |
+
"step": 6000
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 1.105379513633014,
|
| 185 |
+
"eval_loss": 1.1245251893997192,
|
| 186 |
+
"eval_runtime": 74.102,
|
| 187 |
+
"eval_samples_per_second": 16.288,
|
| 188 |
+
"eval_steps_per_second": 2.038,
|
| 189 |
+
"step": 6000
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 1.1974944731024317,
|
| 193 |
+
"grad_norm": 0.00903751328587532,
|
| 194 |
+
"learning_rate": 4.002087939081307e-06,
|
| 195 |
+
"loss": 0.1085,
|
| 196 |
+
"step": 6500
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 1.1974944731024317,
|
| 200 |
+
"eval_loss": 1.1834831237792969,
|
| 201 |
+
"eval_runtime": 74.4532,
|
| 202 |
+
"eval_samples_per_second": 16.212,
|
| 203 |
+
"eval_steps_per_second": 2.028,
|
| 204 |
+
"step": 6500
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 1.2896094325718497,
|
| 208 |
+
"grad_norm": 5.0311258631419165e-11,
|
| 209 |
+
"learning_rate": 3.925325472856792e-06,
|
| 210 |
+
"loss": 0.1174,
|
| 211 |
+
"step": 7000
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 1.2896094325718497,
|
| 215 |
+
"eval_loss": 1.1334316730499268,
|
| 216 |
+
"eval_runtime": 74.1756,
|
| 217 |
+
"eval_samples_per_second": 16.272,
|
| 218 |
+
"eval_steps_per_second": 2.036,
|
| 219 |
+
"step": 7000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 1.3817243920412676,
|
| 223 |
+
"grad_norm": 3.813414917175351e-09,
|
| 224 |
+
"learning_rate": 3.848563006632278e-06,
|
| 225 |
+
"loss": 0.122,
|
| 226 |
+
"step": 7500
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 1.3817243920412676,
|
| 230 |
+
"eval_loss": 1.1058012247085571,
|
| 231 |
+
"eval_runtime": 74.0285,
|
| 232 |
+
"eval_samples_per_second": 16.305,
|
| 233 |
+
"eval_steps_per_second": 2.04,
|
| 234 |
+
"step": 7500
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.4738393515106853,
|
| 238 |
+
"grad_norm": 4.6230165935412515e-06,
|
| 239 |
+
"learning_rate": 3.7718005404077624e-06,
|
| 240 |
+
"loss": 0.1653,
|
| 241 |
+
"step": 8000
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.4738393515106853,
|
| 245 |
+
"eval_loss": 1.8118412494659424,
|
| 246 |
+
"eval_runtime": 73.9465,
|
| 247 |
+
"eval_samples_per_second": 16.323,
|
| 248 |
+
"eval_steps_per_second": 2.042,
|
| 249 |
+
"step": 8000
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 1.565954310980103,
|
| 253 |
+
"grad_norm": 6.103370031596e-10,
|
| 254 |
+
"learning_rate": 3.6950380741832476e-06,
|
| 255 |
+
"loss": 0.1144,
|
| 256 |
+
"step": 8500
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 1.565954310980103,
|
| 260 |
+
"eval_loss": 1.0845656394958496,
|
| 261 |
+
"eval_runtime": 73.8897,
|
| 262 |
+
"eval_samples_per_second": 16.335,
|
| 263 |
+
"eval_steps_per_second": 2.044,
|
| 264 |
+
"step": 8500
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 1.658069270449521,
|
| 268 |
+
"grad_norm": 0.023263949900865555,
|
| 269 |
+
"learning_rate": 3.6182756079587327e-06,
|
| 270 |
+
"loss": 0.1866,
|
| 271 |
+
"step": 9000
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 1.658069270449521,
|
| 275 |
+
"eval_loss": 1.2360613346099854,
|
| 276 |
+
"eval_runtime": 73.8786,
|
| 277 |
+
"eval_samples_per_second": 16.338,
|
| 278 |
+
"eval_steps_per_second": 2.044,
|
| 279 |
+
"step": 9000
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 1.750184229918939,
|
| 283 |
+
"grad_norm": 1.1146563672070897e-09,
|
| 284 |
+
"learning_rate": 3.541513141734218e-06,
|
| 285 |
+
"loss": 0.1385,
|
| 286 |
+
"step": 9500
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 1.750184229918939,
|
| 290 |
+
"eval_loss": 1.0680217742919922,
|
| 291 |
+
"eval_runtime": 73.8064,
|
| 292 |
+
"eval_samples_per_second": 16.354,
|
| 293 |
+
"eval_steps_per_second": 2.046,
|
| 294 |
+
"step": 9500
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 1.8422991893883567,
|
| 298 |
+
"grad_norm": 0.0801646038889885,
|
| 299 |
+
"learning_rate": 3.464750675509703e-06,
|
| 300 |
+
"loss": 0.1178,
|
| 301 |
+
"step": 10000
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 1.8422991893883567,
|
| 305 |
+
"eval_loss": 0.9592632055282593,
|
| 306 |
+
"eval_runtime": 73.9232,
|
| 307 |
+
"eval_samples_per_second": 16.328,
|
| 308 |
+
"eval_steps_per_second": 2.043,
|
| 309 |
+
"step": 10000
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 1.9344141488577744,
|
| 313 |
+
"grad_norm": 8.68252456598384e-08,
|
| 314 |
+
"learning_rate": 3.3879882092851883e-06,
|
| 315 |
+
"loss": 0.1506,
|
| 316 |
+
"step": 10500
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 1.9344141488577744,
|
| 320 |
+
"eval_loss": 1.0038937330245972,
|
| 321 |
+
"eval_runtime": 74.3344,
|
| 322 |
+
"eval_samples_per_second": 16.237,
|
| 323 |
+
"eval_steps_per_second": 2.031,
|
| 324 |
+
"step": 10500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 2.026529108327192,
|
| 328 |
+
"grad_norm": 1.1008462763584248e-07,
|
| 329 |
+
"learning_rate": 3.311225743060673e-06,
|
| 330 |
+
"loss": 0.0938,
|
| 331 |
+
"step": 11000
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 2.026529108327192,
|
| 335 |
+
"eval_loss": 0.931797206401825,
|
| 336 |
+
"eval_runtime": 74.2035,
|
| 337 |
+
"eval_samples_per_second": 16.266,
|
| 338 |
+
"eval_steps_per_second": 2.035,
|
| 339 |
+
"step": 11000
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 2.1186440677966103,
|
| 343 |
+
"grad_norm": 7.901914614194538e-06,
|
| 344 |
+
"learning_rate": 3.234463276836158e-06,
|
| 345 |
+
"loss": 0.093,
|
| 346 |
+
"step": 11500
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 2.1186440677966103,
|
| 350 |
+
"eval_loss": 0.9992227554321289,
|
| 351 |
+
"eval_runtime": 74.2516,
|
| 352 |
+
"eval_samples_per_second": 16.256,
|
| 353 |
+
"eval_steps_per_second": 2.034,
|
| 354 |
+
"step": 11500
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 2.210759027266028,
|
| 358 |
+
"grad_norm": 0.04380282014608383,
|
| 359 |
+
"learning_rate": 3.1577008106116434e-06,
|
| 360 |
+
"loss": 0.0877,
|
| 361 |
+
"step": 12000
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 2.210759027266028,
|
| 365 |
+
"eval_loss": 0.9989096522331238,
|
| 366 |
+
"eval_runtime": 75.2761,
|
| 367 |
+
"eval_samples_per_second": 16.034,
|
| 368 |
+
"eval_steps_per_second": 2.006,
|
| 369 |
+
"step": 12000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 2.3028739867354457,
|
| 373 |
+
"grad_norm": 65.85383605957031,
|
| 374 |
+
"learning_rate": 3.080938344387129e-06,
|
| 375 |
+
"loss": 0.1102,
|
| 376 |
+
"step": 12500
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 2.3028739867354457,
|
| 380 |
+
"eval_loss": 0.9193492531776428,
|
| 381 |
+
"eval_runtime": 74.3653,
|
| 382 |
+
"eval_samples_per_second": 16.231,
|
| 383 |
+
"eval_steps_per_second": 2.031,
|
| 384 |
+
"step": 12500
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 2.3949889462048635,
|
| 388 |
+
"grad_norm": 0.0006995272124186158,
|
| 389 |
+
"learning_rate": 3.004175878162614e-06,
|
| 390 |
+
"loss": 0.0916,
|
| 391 |
+
"step": 13000
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 2.3949889462048635,
|
| 395 |
+
"eval_loss": 0.9769579172134399,
|
| 396 |
+
"eval_runtime": 74.2536,
|
| 397 |
+
"eval_samples_per_second": 16.255,
|
| 398 |
+
"eval_steps_per_second": 2.034,
|
| 399 |
+
"step": 13000
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 2.4871039056742816,
|
| 403 |
+
"grad_norm": 3.747671507881023e-05,
|
| 404 |
+
"learning_rate": 2.9274134119380993e-06,
|
| 405 |
+
"loss": 0.0665,
|
| 406 |
+
"step": 13500
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 2.4871039056742816,
|
| 410 |
+
"eval_loss": 0.982865035533905,
|
| 411 |
+
"eval_runtime": 75.5772,
|
| 412 |
+
"eval_samples_per_second": 15.97,
|
| 413 |
+
"eval_steps_per_second": 1.998,
|
| 414 |
+
"step": 13500
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 2.5792188651436994,
|
| 418 |
+
"grad_norm": 4.0381178855896,
|
| 419 |
+
"learning_rate": 2.850650945713584e-06,
|
| 420 |
+
"loss": 0.0545,
|
| 421 |
+
"step": 14000
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 2.5792188651436994,
|
| 425 |
+
"eval_loss": 0.9746910929679871,
|
| 426 |
+
"eval_runtime": 74.7109,
|
| 427 |
+
"eval_samples_per_second": 16.156,
|
| 428 |
+
"eval_steps_per_second": 2.021,
|
| 429 |
+
"step": 14000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 2.671333824613117,
|
| 433 |
+
"grad_norm": 7.164443260876396e-09,
|
| 434 |
+
"learning_rate": 2.7738884794890693e-06,
|
| 435 |
+
"loss": 0.1023,
|
| 436 |
+
"step": 14500
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 2.671333824613117,
|
| 440 |
+
"eval_loss": 0.9872701168060303,
|
| 441 |
+
"eval_runtime": 73.8157,
|
| 442 |
+
"eval_samples_per_second": 16.352,
|
| 443 |
+
"eval_steps_per_second": 2.046,
|
| 444 |
+
"step": 14500
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 2.7634487840825352,
|
| 448 |
+
"grad_norm": 1.5922936302814605e-09,
|
| 449 |
+
"learning_rate": 2.6971260132645544e-06,
|
| 450 |
+
"loss": 0.1057,
|
| 451 |
+
"step": 15000
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 2.7634487840825352,
|
| 455 |
+
"eval_loss": 0.946104109287262,
|
| 456 |
+
"eval_runtime": 73.9037,
|
| 457 |
+
"eval_samples_per_second": 16.332,
|
| 458 |
+
"eval_steps_per_second": 2.043,
|
| 459 |
+
"step": 15000
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 2.855563743551953,
|
| 463 |
+
"grad_norm": 8.37097544059695e-11,
|
| 464 |
+
"learning_rate": 2.6203635470400396e-06,
|
| 465 |
+
"loss": 0.087,
|
| 466 |
+
"step": 15500
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 2.855563743551953,
|
| 470 |
+
"eval_loss": 0.980368435382843,
|
| 471 |
+
"eval_runtime": 73.8687,
|
| 472 |
+
"eval_samples_per_second": 16.34,
|
| 473 |
+
"eval_steps_per_second": 2.044,
|
| 474 |
+
"step": 15500
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 2.9476787030213707,
|
| 478 |
+
"grad_norm": 1.7481942176818848,
|
| 479 |
+
"learning_rate": 2.543601080815525e-06,
|
| 480 |
+
"loss": 0.046,
|
| 481 |
+
"step": 16000
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 2.9476787030213707,
|
| 485 |
+
"eval_loss": 0.9453437328338623,
|
| 486 |
+
"eval_runtime": 73.8853,
|
| 487 |
+
"eval_samples_per_second": 16.336,
|
| 488 |
+
"eval_steps_per_second": 2.044,
|
| 489 |
+
"step": 16000
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 3.0397936624907884,
|
| 493 |
+
"grad_norm": 2.1040523923848697e-12,
|
| 494 |
+
"learning_rate": 2.46683861459101e-06,
|
| 495 |
+
"loss": 0.0914,
|
| 496 |
+
"step": 16500
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 3.0397936624907884,
|
| 500 |
+
"eval_loss": 0.9565027952194214,
|
| 501 |
+
"eval_runtime": 74.0088,
|
| 502 |
+
"eval_samples_per_second": 16.309,
|
| 503 |
+
"eval_steps_per_second": 2.04,
|
| 504 |
+
"step": 16500
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 3.131908621960206,
|
| 508 |
+
"grad_norm": 1.3199319415235777e-10,
|
| 509 |
+
"learning_rate": 2.390076148366495e-06,
|
| 510 |
+
"loss": 0.0486,
|
| 511 |
+
"step": 17000
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 3.131908621960206,
|
| 515 |
+
"eval_loss": 0.962292492389679,
|
| 516 |
+
"eval_runtime": 73.9793,
|
| 517 |
+
"eval_samples_per_second": 16.315,
|
| 518 |
+
"eval_steps_per_second": 2.041,
|
| 519 |
+
"step": 17000
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 3.2240235814296243,
|
| 523 |
+
"grad_norm": 5.488172246259637e-05,
|
| 524 |
+
"learning_rate": 2.31331368214198e-06,
|
| 525 |
+
"loss": 0.1024,
|
| 526 |
+
"step": 17500
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 3.2240235814296243,
|
| 530 |
+
"eval_loss": 0.8799718022346497,
|
| 531 |
+
"eval_runtime": 73.833,
|
| 532 |
+
"eval_samples_per_second": 16.348,
|
| 533 |
+
"eval_steps_per_second": 2.045,
|
| 534 |
+
"step": 17500
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 3.316138540899042,
|
| 538 |
+
"grad_norm": 1.8391082927393665e-12,
|
| 539 |
+
"learning_rate": 2.236551215917465e-06,
|
| 540 |
+
"loss": 0.0641,
|
| 541 |
+
"step": 18000
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 3.316138540899042,
|
| 545 |
+
"eval_loss": 0.8668168187141418,
|
| 546 |
+
"eval_runtime": 74.1002,
|
| 547 |
+
"eval_samples_per_second": 16.289,
|
| 548 |
+
"eval_steps_per_second": 2.038,
|
| 549 |
+
"step": 18000
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 3.4082535003684598,
|
| 553 |
+
"grad_norm": 4.760212224397264e-10,
|
| 554 |
+
"learning_rate": 2.1597887496929502e-06,
|
| 555 |
+
"loss": 0.0982,
|
| 556 |
+
"step": 18500
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 3.4082535003684598,
|
| 560 |
+
"eval_loss": 0.8307517766952515,
|
| 561 |
+
"eval_runtime": 74.2129,
|
| 562 |
+
"eval_samples_per_second": 16.264,
|
| 563 |
+
"eval_steps_per_second": 2.035,
|
| 564 |
+
"step": 18500
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 3.500368459837878,
|
| 568 |
+
"grad_norm": 0.00010056859900942072,
|
| 569 |
+
"learning_rate": 2.0830262834684354e-06,
|
| 570 |
+
"loss": 0.0256,
|
| 571 |
+
"step": 19000
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 3.500368459837878,
|
| 575 |
+
"eval_loss": 0.7900328636169434,
|
| 576 |
+
"eval_runtime": 73.8942,
|
| 577 |
+
"eval_samples_per_second": 16.334,
|
| 578 |
+
"eval_steps_per_second": 2.043,
|
| 579 |
+
"step": 19000
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 3.5924834193072956,
|
| 583 |
+
"grad_norm": 3.8125975265757006e-07,
|
| 584 |
+
"learning_rate": 2.0062638172439206e-06,
|
| 585 |
+
"loss": 0.0684,
|
| 586 |
+
"step": 19500
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 3.5924834193072956,
|
| 590 |
+
"eval_loss": 0.807513415813446,
|
| 591 |
+
"eval_runtime": 75.2826,
|
| 592 |
+
"eval_samples_per_second": 16.033,
|
| 593 |
+
"eval_steps_per_second": 2.006,
|
| 594 |
+
"step": 19500
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 3.6845983787767134,
|
| 598 |
+
"grad_norm": 2.2643770742464785e-14,
|
| 599 |
+
"learning_rate": 1.9295013510194058e-06,
|
| 600 |
+
"loss": 0.0463,
|
| 601 |
+
"step": 20000
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 3.6845983787767134,
|
| 605 |
+
"eval_loss": 0.7804784178733826,
|
| 606 |
+
"eval_runtime": 74.4513,
|
| 607 |
+
"eval_samples_per_second": 16.212,
|
| 608 |
+
"eval_steps_per_second": 2.028,
|
| 609 |
+
"step": 20000
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 3.776713338246131,
|
| 613 |
+
"grad_norm": 48.942562103271484,
|
| 614 |
+
"learning_rate": 1.8527388847948907e-06,
|
| 615 |
+
"loss": 0.0536,
|
| 616 |
+
"step": 20500
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 3.776713338246131,
|
| 620 |
+
"eval_loss": 0.7935602068901062,
|
| 621 |
+
"eval_runtime": 74.0045,
|
| 622 |
+
"eval_samples_per_second": 16.31,
|
| 623 |
+
"eval_steps_per_second": 2.04,
|
| 624 |
+
"step": 20500
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 3.868828297715549,
|
| 628 |
+
"grad_norm": 1.9078814916806408e-11,
|
| 629 |
+
"learning_rate": 1.775976418570376e-06,
|
| 630 |
+
"loss": 0.0901,
|
| 631 |
+
"step": 21000
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 3.868828297715549,
|
| 635 |
+
"eval_loss": 0.7762666344642639,
|
| 636 |
+
"eval_runtime": 75.3556,
|
| 637 |
+
"eval_samples_per_second": 16.017,
|
| 638 |
+
"eval_steps_per_second": 2.004,
|
| 639 |
+
"step": 21000
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 3.960943257184967,
|
| 643 |
+
"grad_norm": 1.1505435134501353e-16,
|
| 644 |
+
"learning_rate": 1.699213952345861e-06,
|
| 645 |
+
"loss": 0.0299,
|
| 646 |
+
"step": 21500
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 3.960943257184967,
|
| 650 |
+
"eval_loss": 0.8083813786506653,
|
| 651 |
+
"eval_runtime": 74.1656,
|
| 652 |
+
"eval_samples_per_second": 16.274,
|
| 653 |
+
"eval_steps_per_second": 2.036,
|
| 654 |
+
"step": 21500
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 4.053058216654384,
|
| 658 |
+
"grad_norm": 0.4863293766975403,
|
| 659 |
+
"learning_rate": 1.622451486121346e-06,
|
| 660 |
+
"loss": 0.0736,
|
| 661 |
+
"step": 22000
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 4.053058216654384,
|
| 665 |
+
"eval_loss": 0.7665108442306519,
|
| 666 |
+
"eval_runtime": 74.1302,
|
| 667 |
+
"eval_samples_per_second": 16.282,
|
| 668 |
+
"eval_steps_per_second": 2.037,
|
| 669 |
+
"step": 22000
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 4.145173176123802,
|
| 673 |
+
"grad_norm": 3.1499366741627455e-05,
|
| 674 |
+
"learning_rate": 1.5456890198968312e-06,
|
| 675 |
+
"loss": 0.018,
|
| 676 |
+
"step": 22500
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 4.145173176123802,
|
| 680 |
+
"eval_loss": 0.7470127940177917,
|
| 681 |
+
"eval_runtime": 75.4156,
|
| 682 |
+
"eval_samples_per_second": 16.005,
|
| 683 |
+
"eval_steps_per_second": 2.002,
|
| 684 |
+
"step": 22500
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 4.237288135593221,
|
| 688 |
+
"grad_norm": 2.700594468104528e-13,
|
| 689 |
+
"learning_rate": 1.4689265536723166e-06,
|
| 690 |
+
"loss": 0.0409,
|
| 691 |
+
"step": 23000
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 4.237288135593221,
|
| 695 |
+
"eval_loss": 0.751453697681427,
|
| 696 |
+
"eval_runtime": 74.3373,
|
| 697 |
+
"eval_samples_per_second": 16.237,
|
| 698 |
+
"eval_steps_per_second": 2.031,
|
| 699 |
+
"step": 23000
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 4.329403095062638,
|
| 703 |
+
"grad_norm": 9.511566076980671e-07,
|
| 704 |
+
"learning_rate": 1.3921640874478016e-06,
|
| 705 |
+
"loss": 0.0481,
|
| 706 |
+
"step": 23500
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 4.329403095062638,
|
| 710 |
+
"eval_loss": 0.7510703206062317,
|
| 711 |
+
"eval_runtime": 74.0311,
|
| 712 |
+
"eval_samples_per_second": 16.304,
|
| 713 |
+
"eval_steps_per_second": 2.04,
|
| 714 |
+
"step": 23500
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 4.421518054532056,
|
| 718 |
+
"grad_norm": 0.0439174547791481,
|
| 719 |
+
"learning_rate": 1.3154016212232868e-06,
|
| 720 |
+
"loss": 0.0531,
|
| 721 |
+
"step": 24000
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 4.421518054532056,
|
| 725 |
+
"eval_loss": 0.7321518659591675,
|
| 726 |
+
"eval_runtime": 75.3459,
|
| 727 |
+
"eval_samples_per_second": 16.019,
|
| 728 |
+
"eval_steps_per_second": 2.004,
|
| 729 |
+
"step": 24000
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 4.513633014001474,
|
| 733 |
+
"grad_norm": 3.456220032393986e-11,
|
| 734 |
+
"learning_rate": 1.238639154998772e-06,
|
| 735 |
+
"loss": 0.0397,
|
| 736 |
+
"step": 24500
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 4.513633014001474,
|
| 740 |
+
"eval_loss": 0.7277648448944092,
|
| 741 |
+
"eval_runtime": 75.3581,
|
| 742 |
+
"eval_samples_per_second": 16.017,
|
| 743 |
+
"eval_steps_per_second": 2.004,
|
| 744 |
+
"step": 24500
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 4.6057479734708915,
|
| 748 |
+
"grad_norm": 2.3053861752941884e-07,
|
| 749 |
+
"learning_rate": 1.1618766887742571e-06,
|
| 750 |
+
"loss": 0.0244,
|
| 751 |
+
"step": 25000
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 4.6057479734708915,
|
| 755 |
+
"eval_loss": 0.7375216484069824,
|
| 756 |
+
"eval_runtime": 74.2623,
|
| 757 |
+
"eval_samples_per_second": 16.253,
|
| 758 |
+
"eval_steps_per_second": 2.033,
|
| 759 |
+
"step": 25000
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 4.69786293294031,
|
| 763 |
+
"grad_norm": 1.903062374353226e-09,
|
| 764 |
+
"learning_rate": 1.085114222549742e-06,
|
| 765 |
+
"loss": 0.037,
|
| 766 |
+
"step": 25500
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 4.69786293294031,
|
| 770 |
+
"eval_loss": 0.7155526280403137,
|
| 771 |
+
"eval_runtime": 73.9966,
|
| 772 |
+
"eval_samples_per_second": 16.312,
|
| 773 |
+
"eval_steps_per_second": 2.041,
|
| 774 |
+
"step": 25500
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 4.789977892409727,
|
| 778 |
+
"grad_norm": 9.134251514886227e-14,
|
| 779 |
+
"learning_rate": 1.0083517563252273e-06,
|
| 780 |
+
"loss": 0.0289,
|
| 781 |
+
"step": 26000
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 4.789977892409727,
|
| 785 |
+
"eval_loss": 0.7181342244148254,
|
| 786 |
+
"eval_runtime": 75.4204,
|
| 787 |
+
"eval_samples_per_second": 16.004,
|
| 788 |
+
"eval_steps_per_second": 2.002,
|
| 789 |
+
"step": 26000
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 4.882092851879145,
|
| 793 |
+
"grad_norm": 0.07942134141921997,
|
| 794 |
+
"learning_rate": 9.315892901007124e-07,
|
| 795 |
+
"loss": 0.0404,
|
| 796 |
+
"step": 26500
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 4.882092851879145,
|
| 800 |
+
"eval_loss": 0.7185495495796204,
|
| 801 |
+
"eval_runtime": 74.3423,
|
| 802 |
+
"eval_samples_per_second": 16.236,
|
| 803 |
+
"eval_steps_per_second": 2.031,
|
| 804 |
+
"step": 26500
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 4.974207811348563,
|
| 808 |
+
"grad_norm": 3.4462771525411995e-10,
|
| 809 |
+
"learning_rate": 8.548268238761975e-07,
|
| 810 |
+
"loss": 0.0281,
|
| 811 |
+
"step": 27000
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 4.974207811348563,
|
| 815 |
+
"eval_loss": 0.7134640216827393,
|
| 816 |
+
"eval_runtime": 74.0291,
|
| 817 |
+
"eval_samples_per_second": 16.304,
|
| 818 |
+
"eval_steps_per_second": 2.04,
|
| 819 |
+
"step": 27000
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 5.0663227708179805,
|
| 823 |
+
"grad_norm": 2.4655478227941785e-06,
|
| 824 |
+
"learning_rate": 7.780643576516828e-07,
|
| 825 |
+
"loss": 0.0377,
|
| 826 |
+
"step": 27500
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 5.0663227708179805,
|
| 830 |
+
"eval_loss": 0.6949731111526489,
|
| 831 |
+
"eval_runtime": 73.7561,
|
| 832 |
+
"eval_samples_per_second": 16.365,
|
| 833 |
+
"eval_steps_per_second": 2.047,
|
| 834 |
+
"step": 27500
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 5.158437730287399,
|
| 838 |
+
"grad_norm": 0.0001618898386368528,
|
| 839 |
+
"learning_rate": 7.013018914271679e-07,
|
| 840 |
+
"loss": 0.0428,
|
| 841 |
+
"step": 28000
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 5.158437730287399,
|
| 845 |
+
"eval_loss": 0.6796632409095764,
|
| 846 |
+
"eval_runtime": 75.8163,
|
| 847 |
+
"eval_samples_per_second": 15.92,
|
| 848 |
+
"eval_steps_per_second": 1.992,
|
| 849 |
+
"step": 28000
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 5.250552689756817,
|
| 853 |
+
"grad_norm": 9.771591976459604e-06,
|
| 854 |
+
"learning_rate": 6.24539425202653e-07,
|
| 855 |
+
"loss": 0.0309,
|
| 856 |
+
"step": 28500
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 5.250552689756817,
|
| 860 |
+
"eval_loss": 0.6769527792930603,
|
| 861 |
+
"eval_runtime": 73.9476,
|
| 862 |
+
"eval_samples_per_second": 16.322,
|
| 863 |
+
"eval_steps_per_second": 2.042,
|
| 864 |
+
"step": 28500
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 5.342667649226234,
|
| 868 |
+
"grad_norm": 1.635995129593768e-21,
|
| 869 |
+
"learning_rate": 5.477769589781381e-07,
|
| 870 |
+
"loss": 0.051,
|
| 871 |
+
"step": 29000
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 5.342667649226234,
|
| 875 |
+
"eval_loss": 0.6774747967720032,
|
| 876 |
+
"eval_runtime": 74.0097,
|
| 877 |
+
"eval_samples_per_second": 16.309,
|
| 878 |
+
"eval_steps_per_second": 2.04,
|
| 879 |
+
"step": 29000
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 5.434782608695652,
|
| 883 |
+
"grad_norm": 0.3365604281425476,
|
| 884 |
+
"learning_rate": 4.7101449275362324e-07,
|
| 885 |
+
"loss": 0.0207,
|
| 886 |
+
"step": 29500
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 5.434782608695652,
|
| 890 |
+
"eval_loss": 0.660186231136322,
|
| 891 |
+
"eval_runtime": 75.7037,
|
| 892 |
+
"eval_samples_per_second": 15.944,
|
| 893 |
+
"eval_steps_per_second": 1.995,
|
| 894 |
+
"step": 29500
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 5.5268975681650705,
|
| 898 |
+
"grad_norm": 0.0016033226856961846,
|
| 899 |
+
"learning_rate": 3.942520265291084e-07,
|
| 900 |
+
"loss": 0.0124,
|
| 901 |
+
"step": 30000
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 5.5268975681650705,
|
| 905 |
+
"eval_loss": 0.6515346765518188,
|
| 906 |
+
"eval_runtime": 74.0515,
|
| 907 |
+
"eval_samples_per_second": 16.299,
|
| 908 |
+
"eval_steps_per_second": 2.039,
|
| 909 |
+
"step": 30000
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 5.619012527634488,
|
| 913 |
+
"grad_norm": 3.3180751479733317e-09,
|
| 914 |
+
"learning_rate": 3.174895603045935e-07,
|
| 915 |
+
"loss": 0.0229,
|
| 916 |
+
"step": 30500
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 5.619012527634488,
|
| 920 |
+
"eval_loss": 0.6515493988990784,
|
| 921 |
+
"eval_runtime": 73.9272,
|
| 922 |
+
"eval_samples_per_second": 16.327,
|
| 923 |
+
"eval_steps_per_second": 2.043,
|
| 924 |
+
"step": 30500
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 5.711127487103906,
|
| 928 |
+
"grad_norm": 1.4150446222757629e-12,
|
| 929 |
+
"learning_rate": 2.407270940800786e-07,
|
| 930 |
+
"loss": 0.022,
|
| 931 |
+
"step": 31000
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 5.711127487103906,
|
| 935 |
+
"eval_loss": 0.6436796188354492,
|
| 936 |
+
"eval_runtime": 75.582,
|
| 937 |
+
"eval_samples_per_second": 15.969,
|
| 938 |
+
"eval_steps_per_second": 1.998,
|
| 939 |
+
"step": 31000
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 5.803242446573323,
|
| 943 |
+
"grad_norm": 7.415730529647391e-11,
|
| 944 |
+
"learning_rate": 1.6396462785556376e-07,
|
| 945 |
+
"loss": 0.0328,
|
| 946 |
+
"step": 31500
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 5.803242446573323,
|
| 950 |
+
"eval_loss": 0.6397524476051331,
|
| 951 |
+
"eval_runtime": 73.7673,
|
| 952 |
+
"eval_samples_per_second": 16.362,
|
| 953 |
+
"eval_steps_per_second": 2.047,
|
| 954 |
+
"step": 31500
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 5.895357406042741,
|
| 958 |
+
"grad_norm": 5.1268246592517386e-11,
|
| 959 |
+
"learning_rate": 8.72021616310489e-08,
|
| 960 |
+
"loss": 0.0233,
|
| 961 |
+
"step": 32000
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 5.895357406042741,
|
| 965 |
+
"eval_loss": 0.6386470794677734,
|
| 966 |
+
"eval_runtime": 73.9193,
|
| 967 |
+
"eval_samples_per_second": 16.329,
|
| 968 |
+
"eval_steps_per_second": 2.043,
|
| 969 |
+
"step": 32000
|
| 970 |
+
}
|
| 971 |
+
],
|
| 972 |
+
"logging_steps": 500,
|
| 973 |
+
"max_steps": 32568,
|
| 974 |
+
"num_input_tokens_seen": 0,
|
| 975 |
+
"num_train_epochs": 6,
|
| 976 |
+
"save_steps": 500,
|
| 977 |
+
"stateful_callbacks": {
|
| 978 |
+
"TrainerControl": {
|
| 979 |
+
"args": {
|
| 980 |
+
"should_epoch_stop": false,
|
| 981 |
+
"should_evaluate": false,
|
| 982 |
+
"should_log": false,
|
| 983 |
+
"should_save": true,
|
| 984 |
+
"should_training_stop": false
|
| 985 |
+
},
|
| 986 |
+
"attributes": {}
|
| 987 |
+
}
|
| 988 |
+
},
|
| 989 |
+
"total_flos": 1.151298675702252e+16,
|
| 990 |
+
"train_batch_size": 2,
|
| 991 |
+
"trial_name": null,
|
| 992 |
+
"trial_params": null
|
| 993 |
+
}
|
checkpoint-32000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:09987e134b23188d490a3f3d609c84b1bd02ce771791cf447dcb77366fe93878
|
| 3 |
+
size 4847
|