Upload 12 files
Browse files- .gitattributes +1 -0
- all_results.json +13 -0
- config.json +52 -0
- eval_results.json +8 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- preprocessor_config.json +36 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- state.db +3 -0
- train_results.json +8 -0
- trainer_state.json +1758 -0
- training_args.bin +3 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
state.db filter=lfs diff=lfs merge=lfs -text
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 20.0,
|
| 3 |
+
"eval_accuracy": 0.7354497354497355,
|
| 4 |
+
"eval_loss": 0.8083848357200623,
|
| 5 |
+
"eval_runtime": 1.8368,
|
| 6 |
+
"eval_samples_per_second": 102.896,
|
| 7 |
+
"eval_steps_per_second": 6.533,
|
| 8 |
+
"total_flos": 2.681093741830963e+18,
|
| 9 |
+
"train_loss": 0.29336747460895113,
|
| 10 |
+
"train_runtime": 795.2059,
|
| 11 |
+
"train_samples_per_second": 42.681,
|
| 12 |
+
"train_steps_per_second": 1.358
|
| 13 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,52 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "google/vit-base-patch32-224-in21k",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"ViTForImageClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.0,
|
| 7 |
+
"encoder_stride": 16,
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.0,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"id2label": {
|
| 12 |
+
"0": "afro",
|
| 13 |
+
"1": "classical",
|
| 14 |
+
"10": "reggae",
|
| 15 |
+
"11": "rock",
|
| 16 |
+
"2": "country",
|
| 17 |
+
"3": "disco",
|
| 18 |
+
"4": "electro",
|
| 19 |
+
"5": "jazz",
|
| 20 |
+
"6": "latin",
|
| 21 |
+
"7": "metal",
|
| 22 |
+
"8": "pop",
|
| 23 |
+
"9": "rap"
|
| 24 |
+
},
|
| 25 |
+
"image_size": 224,
|
| 26 |
+
"initializer_range": 0.02,
|
| 27 |
+
"intermediate_size": 3072,
|
| 28 |
+
"label2id": {
|
| 29 |
+
"afro": "0",
|
| 30 |
+
"classical": "1",
|
| 31 |
+
"country": "2",
|
| 32 |
+
"disco": "3",
|
| 33 |
+
"electro": "4",
|
| 34 |
+
"jazz": "5",
|
| 35 |
+
"latin": "6",
|
| 36 |
+
"metal": "7",
|
| 37 |
+
"pop": "8",
|
| 38 |
+
"rap": "9",
|
| 39 |
+
"reggae": "10",
|
| 40 |
+
"rock": "11"
|
| 41 |
+
},
|
| 42 |
+
"layer_norm_eps": 1e-12,
|
| 43 |
+
"model_type": "vit",
|
| 44 |
+
"num_attention_heads": 12,
|
| 45 |
+
"num_channels": 3,
|
| 46 |
+
"num_hidden_layers": 12,
|
| 47 |
+
"patch_size": 32,
|
| 48 |
+
"problem_type": "single_label_classification",
|
| 49 |
+
"qkv_bias": true,
|
| 50 |
+
"torch_dtype": "float32",
|
| 51 |
+
"transformers_version": "4.39.3"
|
| 52 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 20.0,
|
| 3 |
+
"eval_accuracy": 0.7354497354497355,
|
| 4 |
+
"eval_loss": 0.8083848357200623,
|
| 5 |
+
"eval_runtime": 1.8368,
|
| 6 |
+
"eval_samples_per_second": 102.896,
|
| 7 |
+
"eval_steps_per_second": 6.533
|
| 8 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c8725ef095ab642a6726508ccb349c58f1cf4b687bfbe4891c0c80de6b55bf34
|
| 3 |
+
size 349881056
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a0ea76f7d71a48349ddcc05f83ddbf8f8f117f78a49c05dd6c38d1e6db47a9d1
|
| 3 |
+
size 699882938
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_valid_processor_keys": [
|
| 3 |
+
"images",
|
| 4 |
+
"do_resize",
|
| 5 |
+
"size",
|
| 6 |
+
"resample",
|
| 7 |
+
"do_rescale",
|
| 8 |
+
"rescale_factor",
|
| 9 |
+
"do_normalize",
|
| 10 |
+
"image_mean",
|
| 11 |
+
"image_std",
|
| 12 |
+
"return_tensors",
|
| 13 |
+
"data_format",
|
| 14 |
+
"input_data_format"
|
| 15 |
+
],
|
| 16 |
+
"do_normalize": true,
|
| 17 |
+
"do_rescale": true,
|
| 18 |
+
"do_resize": true,
|
| 19 |
+
"image_mean": [
|
| 20 |
+
0.5,
|
| 21 |
+
0.5,
|
| 22 |
+
0.5
|
| 23 |
+
],
|
| 24 |
+
"image_processor_type": "ViTImageProcessor",
|
| 25 |
+
"image_std": [
|
| 26 |
+
0.5,
|
| 27 |
+
0.5,
|
| 28 |
+
0.5
|
| 29 |
+
],
|
| 30 |
+
"resample": 2,
|
| 31 |
+
"rescale_factor": 0.00392156862745098,
|
| 32 |
+
"size": {
|
| 33 |
+
"height": 224,
|
| 34 |
+
"width": 224
|
| 35 |
+
}
|
| 36 |
+
}
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b3bd3cafcd141485c5526689e7070ba65dab1e4639fbae44141ae41439003c1f
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9022742c9d6746d087ef37f0ee553659edde91d2108327185e4b66eabe4bca26
|
| 3 |
+
size 1064
|
state.db
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cd10d354f3aa4aec0dae3940ee299145d3d33a6cd74f1f01704018e2903f2f3b
|
| 3 |
+
size 349940282
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 20.0,
|
| 3 |
+
"total_flos": 2.681093741830963e+18,
|
| 4 |
+
"train_loss": 0.29336747460895113,
|
| 5 |
+
"train_runtime": 795.2059,
|
| 6 |
+
"train_samples_per_second": 42.681,
|
| 7 |
+
"train_steps_per_second": 1.358
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1758 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8083848357200623,
|
| 3 |
+
"best_model_checkpoint": "./kaggle/working/eGTZANplus/checkpoint-220",
|
| 4 |
+
"epoch": 20.0,
|
| 5 |
+
"eval_steps": 10,
|
| 6 |
+
"global_step": 1080,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.19,
|
| 13 |
+
"grad_norm": 1.4207828044891357,
|
| 14 |
+
"learning_rate": 0.00019814814814814814,
|
| 15 |
+
"loss": 2.4003,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.19,
|
| 20 |
+
"eval_accuracy": 0.19576719576719576,
|
| 21 |
+
"eval_loss": 2.282846689224243,
|
| 22 |
+
"eval_runtime": 1.8415,
|
| 23 |
+
"eval_samples_per_second": 102.636,
|
| 24 |
+
"eval_steps_per_second": 6.517,
|
| 25 |
+
"step": 10
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.37,
|
| 29 |
+
"grad_norm": 1.2310184240341187,
|
| 30 |
+
"learning_rate": 0.0001962962962962963,
|
| 31 |
+
"loss": 2.1703,
|
| 32 |
+
"step": 20
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.37,
|
| 36 |
+
"eval_accuracy": 0.35978835978835977,
|
| 37 |
+
"eval_loss": 1.9852432012557983,
|
| 38 |
+
"eval_runtime": 1.793,
|
| 39 |
+
"eval_samples_per_second": 105.41,
|
| 40 |
+
"eval_steps_per_second": 6.693,
|
| 41 |
+
"step": 20
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 0.56,
|
| 45 |
+
"grad_norm": 2.0627870559692383,
|
| 46 |
+
"learning_rate": 0.00019444444444444446,
|
| 47 |
+
"loss": 1.9696,
|
| 48 |
+
"step": 30
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 0.56,
|
| 52 |
+
"eval_accuracy": 0.3915343915343915,
|
| 53 |
+
"eval_loss": 1.8232808113098145,
|
| 54 |
+
"eval_runtime": 1.786,
|
| 55 |
+
"eval_samples_per_second": 105.821,
|
| 56 |
+
"eval_steps_per_second": 6.719,
|
| 57 |
+
"step": 30
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.74,
|
| 61 |
+
"grad_norm": 1.5207180976867676,
|
| 62 |
+
"learning_rate": 0.0001925925925925926,
|
| 63 |
+
"loss": 1.8051,
|
| 64 |
+
"step": 40
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.74,
|
| 68 |
+
"eval_accuracy": 0.48677248677248675,
|
| 69 |
+
"eval_loss": 1.6591798067092896,
|
| 70 |
+
"eval_runtime": 1.7501,
|
| 71 |
+
"eval_samples_per_second": 107.997,
|
| 72 |
+
"eval_steps_per_second": 6.857,
|
| 73 |
+
"step": 40
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.93,
|
| 77 |
+
"grad_norm": 2.221734046936035,
|
| 78 |
+
"learning_rate": 0.00019074074074074075,
|
| 79 |
+
"loss": 1.6692,
|
| 80 |
+
"step": 50
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.93,
|
| 84 |
+
"eval_accuracy": 0.582010582010582,
|
| 85 |
+
"eval_loss": 1.5287415981292725,
|
| 86 |
+
"eval_runtime": 1.7993,
|
| 87 |
+
"eval_samples_per_second": 105.039,
|
| 88 |
+
"eval_steps_per_second": 6.669,
|
| 89 |
+
"step": 50
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 1.11,
|
| 93 |
+
"grad_norm": 1.5369292497634888,
|
| 94 |
+
"learning_rate": 0.0001890740740740741,
|
| 95 |
+
"loss": 1.5283,
|
| 96 |
+
"step": 60
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 1.11,
|
| 100 |
+
"eval_accuracy": 0.5608465608465608,
|
| 101 |
+
"eval_loss": 1.4252889156341553,
|
| 102 |
+
"eval_runtime": 1.7582,
|
| 103 |
+
"eval_samples_per_second": 107.493,
|
| 104 |
+
"eval_steps_per_second": 6.825,
|
| 105 |
+
"step": 60
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 1.3,
|
| 109 |
+
"grad_norm": 1.9959373474121094,
|
| 110 |
+
"learning_rate": 0.00018722222222222222,
|
| 111 |
+
"loss": 1.3981,
|
| 112 |
+
"step": 70
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 1.3,
|
| 116 |
+
"eval_accuracy": 0.5925925925925926,
|
| 117 |
+
"eval_loss": 1.3883891105651855,
|
| 118 |
+
"eval_runtime": 1.7749,
|
| 119 |
+
"eval_samples_per_second": 106.485,
|
| 120 |
+
"eval_steps_per_second": 6.761,
|
| 121 |
+
"step": 70
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 1.48,
|
| 125 |
+
"grad_norm": 2.101576805114746,
|
| 126 |
+
"learning_rate": 0.00018537037037037038,
|
| 127 |
+
"loss": 1.3047,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 1.48,
|
| 132 |
+
"eval_accuracy": 0.5767195767195767,
|
| 133 |
+
"eval_loss": 1.356843113899231,
|
| 134 |
+
"eval_runtime": 1.7875,
|
| 135 |
+
"eval_samples_per_second": 105.735,
|
| 136 |
+
"eval_steps_per_second": 6.713,
|
| 137 |
+
"step": 80
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 1.67,
|
| 141 |
+
"grad_norm": 1.9240992069244385,
|
| 142 |
+
"learning_rate": 0.00018351851851851854,
|
| 143 |
+
"loss": 1.1325,
|
| 144 |
+
"step": 90
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 1.67,
|
| 148 |
+
"eval_accuracy": 0.6349206349206349,
|
| 149 |
+
"eval_loss": 1.2104465961456299,
|
| 150 |
+
"eval_runtime": 1.7741,
|
| 151 |
+
"eval_samples_per_second": 106.533,
|
| 152 |
+
"eval_steps_per_second": 6.764,
|
| 153 |
+
"step": 90
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 1.85,
|
| 157 |
+
"grad_norm": 1.6294556856155396,
|
| 158 |
+
"learning_rate": 0.00018166666666666667,
|
| 159 |
+
"loss": 1.2004,
|
| 160 |
+
"step": 100
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 1.85,
|
| 164 |
+
"eval_accuracy": 0.6137566137566137,
|
| 165 |
+
"eval_loss": 1.263272762298584,
|
| 166 |
+
"eval_runtime": 1.8419,
|
| 167 |
+
"eval_samples_per_second": 102.609,
|
| 168 |
+
"eval_steps_per_second": 6.515,
|
| 169 |
+
"step": 100
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 2.04,
|
| 173 |
+
"grad_norm": 4.842734336853027,
|
| 174 |
+
"learning_rate": 0.0001798148148148148,
|
| 175 |
+
"loss": 1.0475,
|
| 176 |
+
"step": 110
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 2.04,
|
| 180 |
+
"eval_accuracy": 0.5555555555555556,
|
| 181 |
+
"eval_loss": 1.3616496324539185,
|
| 182 |
+
"eval_runtime": 1.7824,
|
| 183 |
+
"eval_samples_per_second": 106.036,
|
| 184 |
+
"eval_steps_per_second": 6.732,
|
| 185 |
+
"step": 110
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 2.22,
|
| 189 |
+
"grad_norm": 1.8519538640975952,
|
| 190 |
+
"learning_rate": 0.00017796296296296296,
|
| 191 |
+
"loss": 0.9801,
|
| 192 |
+
"step": 120
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 2.22,
|
| 196 |
+
"eval_accuracy": 0.671957671957672,
|
| 197 |
+
"eval_loss": 1.1471754312515259,
|
| 198 |
+
"eval_runtime": 1.796,
|
| 199 |
+
"eval_samples_per_second": 105.234,
|
| 200 |
+
"eval_steps_per_second": 6.682,
|
| 201 |
+
"step": 120
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 2.41,
|
| 205 |
+
"grad_norm": 3.018026351928711,
|
| 206 |
+
"learning_rate": 0.00017611111111111112,
|
| 207 |
+
"loss": 0.862,
|
| 208 |
+
"step": 130
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 2.41,
|
| 212 |
+
"eval_accuracy": 0.6984126984126984,
|
| 213 |
+
"eval_loss": 1.0452642440795898,
|
| 214 |
+
"eval_runtime": 1.7578,
|
| 215 |
+
"eval_samples_per_second": 107.521,
|
| 216 |
+
"eval_steps_per_second": 6.827,
|
| 217 |
+
"step": 130
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 2.59,
|
| 221 |
+
"grad_norm": 2.8672127723693848,
|
| 222 |
+
"learning_rate": 0.00017425925925925928,
|
| 223 |
+
"loss": 0.8905,
|
| 224 |
+
"step": 140
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 2.59,
|
| 228 |
+
"eval_accuracy": 0.6825396825396826,
|
| 229 |
+
"eval_loss": 0.9718140363693237,
|
| 230 |
+
"eval_runtime": 1.8323,
|
| 231 |
+
"eval_samples_per_second": 103.148,
|
| 232 |
+
"eval_steps_per_second": 6.549,
|
| 233 |
+
"step": 140
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 2.78,
|
| 237 |
+
"grad_norm": 3.5106003284454346,
|
| 238 |
+
"learning_rate": 0.00017240740740740742,
|
| 239 |
+
"loss": 0.7839,
|
| 240 |
+
"step": 150
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 2.78,
|
| 244 |
+
"eval_accuracy": 0.6666666666666666,
|
| 245 |
+
"eval_loss": 1.0531541109085083,
|
| 246 |
+
"eval_runtime": 1.7655,
|
| 247 |
+
"eval_samples_per_second": 107.049,
|
| 248 |
+
"eval_steps_per_second": 6.797,
|
| 249 |
+
"step": 150
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 2.96,
|
| 253 |
+
"grad_norm": 2.7532589435577393,
|
| 254 |
+
"learning_rate": 0.00017055555555555555,
|
| 255 |
+
"loss": 0.8304,
|
| 256 |
+
"step": 160
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 2.96,
|
| 260 |
+
"eval_accuracy": 0.6878306878306878,
|
| 261 |
+
"eval_loss": 0.96842360496521,
|
| 262 |
+
"eval_runtime": 1.8371,
|
| 263 |
+
"eval_samples_per_second": 102.881,
|
| 264 |
+
"eval_steps_per_second": 6.532,
|
| 265 |
+
"step": 160
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 3.15,
|
| 269 |
+
"grad_norm": 2.1222331523895264,
|
| 270 |
+
"learning_rate": 0.0001687037037037037,
|
| 271 |
+
"loss": 0.883,
|
| 272 |
+
"step": 170
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 3.15,
|
| 276 |
+
"eval_accuracy": 0.6931216931216931,
|
| 277 |
+
"eval_loss": 0.9298208951950073,
|
| 278 |
+
"eval_runtime": 1.7867,
|
| 279 |
+
"eval_samples_per_second": 105.782,
|
| 280 |
+
"eval_steps_per_second": 6.716,
|
| 281 |
+
"step": 170
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 3.33,
|
| 285 |
+
"grad_norm": 2.5858914852142334,
|
| 286 |
+
"learning_rate": 0.00016685185185185187,
|
| 287 |
+
"loss": 0.5714,
|
| 288 |
+
"step": 180
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 3.33,
|
| 292 |
+
"eval_accuracy": 0.6772486772486772,
|
| 293 |
+
"eval_loss": 0.9491019248962402,
|
| 294 |
+
"eval_runtime": 1.7856,
|
| 295 |
+
"eval_samples_per_second": 105.846,
|
| 296 |
+
"eval_steps_per_second": 6.72,
|
| 297 |
+
"step": 180
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 3.52,
|
| 301 |
+
"grad_norm": 1.7296024560928345,
|
| 302 |
+
"learning_rate": 0.000165,
|
| 303 |
+
"loss": 0.5209,
|
| 304 |
+
"step": 190
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 3.52,
|
| 308 |
+
"eval_accuracy": 0.6984126984126984,
|
| 309 |
+
"eval_loss": 0.914806604385376,
|
| 310 |
+
"eval_runtime": 1.7453,
|
| 311 |
+
"eval_samples_per_second": 108.289,
|
| 312 |
+
"eval_steps_per_second": 6.875,
|
| 313 |
+
"step": 190
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 3.7,
|
| 317 |
+
"grad_norm": 4.235101699829102,
|
| 318 |
+
"learning_rate": 0.00016314814814814816,
|
| 319 |
+
"loss": 0.5404,
|
| 320 |
+
"step": 200
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 3.7,
|
| 324 |
+
"eval_accuracy": 0.671957671957672,
|
| 325 |
+
"eval_loss": 1.0290465354919434,
|
| 326 |
+
"eval_runtime": 1.8123,
|
| 327 |
+
"eval_samples_per_second": 104.285,
|
| 328 |
+
"eval_steps_per_second": 6.621,
|
| 329 |
+
"step": 200
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 3.89,
|
| 333 |
+
"grad_norm": 3.8817615509033203,
|
| 334 |
+
"learning_rate": 0.0001612962962962963,
|
| 335 |
+
"loss": 0.6133,
|
| 336 |
+
"step": 210
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 3.89,
|
| 340 |
+
"eval_accuracy": 0.7142857142857143,
|
| 341 |
+
"eval_loss": 0.9116460680961609,
|
| 342 |
+
"eval_runtime": 1.7735,
|
| 343 |
+
"eval_samples_per_second": 106.57,
|
| 344 |
+
"eval_steps_per_second": 6.766,
|
| 345 |
+
"step": 210
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 4.07,
|
| 349 |
+
"grad_norm": 1.743445634841919,
|
| 350 |
+
"learning_rate": 0.00015944444444444445,
|
| 351 |
+
"loss": 0.4347,
|
| 352 |
+
"step": 220
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 4.07,
|
| 356 |
+
"eval_accuracy": 0.7354497354497355,
|
| 357 |
+
"eval_loss": 0.8083848357200623,
|
| 358 |
+
"eval_runtime": 1.8193,
|
| 359 |
+
"eval_samples_per_second": 103.884,
|
| 360 |
+
"eval_steps_per_second": 6.596,
|
| 361 |
+
"step": 220
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 4.26,
|
| 365 |
+
"grad_norm": 1.8867310285568237,
|
| 366 |
+
"learning_rate": 0.0001575925925925926,
|
| 367 |
+
"loss": 0.3659,
|
| 368 |
+
"step": 230
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 4.26,
|
| 372 |
+
"eval_accuracy": 0.7142857142857143,
|
| 373 |
+
"eval_loss": 0.890904426574707,
|
| 374 |
+
"eval_runtime": 1.7392,
|
| 375 |
+
"eval_samples_per_second": 108.672,
|
| 376 |
+
"eval_steps_per_second": 6.9,
|
| 377 |
+
"step": 230
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 4.44,
|
| 381 |
+
"grad_norm": 2.56878399848938,
|
| 382 |
+
"learning_rate": 0.00015574074074074074,
|
| 383 |
+
"loss": 0.4439,
|
| 384 |
+
"step": 240
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 4.44,
|
| 388 |
+
"eval_accuracy": 0.6825396825396826,
|
| 389 |
+
"eval_loss": 0.9554860591888428,
|
| 390 |
+
"eval_runtime": 1.7559,
|
| 391 |
+
"eval_samples_per_second": 107.64,
|
| 392 |
+
"eval_steps_per_second": 6.834,
|
| 393 |
+
"step": 240
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 4.63,
|
| 397 |
+
"grad_norm": 1.9487425088882446,
|
| 398 |
+
"learning_rate": 0.0001538888888888889,
|
| 399 |
+
"loss": 0.3335,
|
| 400 |
+
"step": 250
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 4.63,
|
| 404 |
+
"eval_accuracy": 0.708994708994709,
|
| 405 |
+
"eval_loss": 0.931969404220581,
|
| 406 |
+
"eval_runtime": 1.8636,
|
| 407 |
+
"eval_samples_per_second": 101.417,
|
| 408 |
+
"eval_steps_per_second": 6.439,
|
| 409 |
+
"step": 250
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 4.81,
|
| 413 |
+
"grad_norm": 2.4911906719207764,
|
| 414 |
+
"learning_rate": 0.00015203703703703703,
|
| 415 |
+
"loss": 0.3695,
|
| 416 |
+
"step": 260
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 4.81,
|
| 420 |
+
"eval_accuracy": 0.7037037037037037,
|
| 421 |
+
"eval_loss": 0.9643996357917786,
|
| 422 |
+
"eval_runtime": 1.743,
|
| 423 |
+
"eval_samples_per_second": 108.437,
|
| 424 |
+
"eval_steps_per_second": 6.885,
|
| 425 |
+
"step": 260
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 5.0,
|
| 429 |
+
"grad_norm": 0.4799601137638092,
|
| 430 |
+
"learning_rate": 0.0001501851851851852,
|
| 431 |
+
"loss": 0.3018,
|
| 432 |
+
"step": 270
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 5.0,
|
| 436 |
+
"eval_accuracy": 0.6455026455026455,
|
| 437 |
+
"eval_loss": 1.1127641201019287,
|
| 438 |
+
"eval_runtime": 1.8057,
|
| 439 |
+
"eval_samples_per_second": 104.667,
|
| 440 |
+
"eval_steps_per_second": 6.646,
|
| 441 |
+
"step": 270
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 5.19,
|
| 445 |
+
"grad_norm": 0.8545930981636047,
|
| 446 |
+
"learning_rate": 0.00014833333333333335,
|
| 447 |
+
"loss": 0.2418,
|
| 448 |
+
"step": 280
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 5.19,
|
| 452 |
+
"eval_accuracy": 0.7301587301587301,
|
| 453 |
+
"eval_loss": 0.8752605319023132,
|
| 454 |
+
"eval_runtime": 1.7714,
|
| 455 |
+
"eval_samples_per_second": 106.698,
|
| 456 |
+
"eval_steps_per_second": 6.774,
|
| 457 |
+
"step": 280
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 5.37,
|
| 461 |
+
"grad_norm": 2.0490822792053223,
|
| 462 |
+
"learning_rate": 0.00014648148148148148,
|
| 463 |
+
"loss": 0.2305,
|
| 464 |
+
"step": 290
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 5.37,
|
| 468 |
+
"eval_accuracy": 0.7142857142857143,
|
| 469 |
+
"eval_loss": 0.9517038464546204,
|
| 470 |
+
"eval_runtime": 1.7422,
|
| 471 |
+
"eval_samples_per_second": 108.483,
|
| 472 |
+
"eval_steps_per_second": 6.888,
|
| 473 |
+
"step": 290
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 5.56,
|
| 477 |
+
"grad_norm": 1.5348315238952637,
|
| 478 |
+
"learning_rate": 0.00014462962962962962,
|
| 479 |
+
"loss": 0.238,
|
| 480 |
+
"step": 300
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 5.56,
|
| 484 |
+
"eval_accuracy": 0.7248677248677249,
|
| 485 |
+
"eval_loss": 0.9478802680969238,
|
| 486 |
+
"eval_runtime": 1.7999,
|
| 487 |
+
"eval_samples_per_second": 105.006,
|
| 488 |
+
"eval_steps_per_second": 6.667,
|
| 489 |
+
"step": 300
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 5.74,
|
| 493 |
+
"grad_norm": 2.6169273853302,
|
| 494 |
+
"learning_rate": 0.00014277777777777778,
|
| 495 |
+
"loss": 0.2099,
|
| 496 |
+
"step": 310
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 5.74,
|
| 500 |
+
"eval_accuracy": 0.671957671957672,
|
| 501 |
+
"eval_loss": 1.103389024734497,
|
| 502 |
+
"eval_runtime": 1.8453,
|
| 503 |
+
"eval_samples_per_second": 102.42,
|
| 504 |
+
"eval_steps_per_second": 6.503,
|
| 505 |
+
"step": 310
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 5.93,
|
| 509 |
+
"grad_norm": 2.5781023502349854,
|
| 510 |
+
"learning_rate": 0.00014092592592592594,
|
| 511 |
+
"loss": 0.2284,
|
| 512 |
+
"step": 320
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 5.93,
|
| 516 |
+
"eval_accuracy": 0.6825396825396826,
|
| 517 |
+
"eval_loss": 1.031624674797058,
|
| 518 |
+
"eval_runtime": 1.7579,
|
| 519 |
+
"eval_samples_per_second": 107.517,
|
| 520 |
+
"eval_steps_per_second": 6.826,
|
| 521 |
+
"step": 320
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 6.11,
|
| 525 |
+
"grad_norm": 3.042239189147949,
|
| 526 |
+
"learning_rate": 0.0001390740740740741,
|
| 527 |
+
"loss": 0.1694,
|
| 528 |
+
"step": 330
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 6.11,
|
| 532 |
+
"eval_accuracy": 0.6613756613756614,
|
| 533 |
+
"eval_loss": 1.1174468994140625,
|
| 534 |
+
"eval_runtime": 1.7854,
|
| 535 |
+
"eval_samples_per_second": 105.856,
|
| 536 |
+
"eval_steps_per_second": 6.721,
|
| 537 |
+
"step": 330
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 6.3,
|
| 541 |
+
"grad_norm": 0.8211657404899597,
|
| 542 |
+
"learning_rate": 0.00013722222222222223,
|
| 543 |
+
"loss": 0.1715,
|
| 544 |
+
"step": 340
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 6.3,
|
| 548 |
+
"eval_accuracy": 0.6772486772486772,
|
| 549 |
+
"eval_loss": 1.1067023277282715,
|
| 550 |
+
"eval_runtime": 1.8157,
|
| 551 |
+
"eval_samples_per_second": 104.091,
|
| 552 |
+
"eval_steps_per_second": 6.609,
|
| 553 |
+
"step": 340
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 6.48,
|
| 557 |
+
"grad_norm": 1.9425742626190186,
|
| 558 |
+
"learning_rate": 0.00013537037037037036,
|
| 559 |
+
"loss": 0.123,
|
| 560 |
+
"step": 350
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 6.48,
|
| 564 |
+
"eval_accuracy": 0.7142857142857143,
|
| 565 |
+
"eval_loss": 1.0037899017333984,
|
| 566 |
+
"eval_runtime": 1.786,
|
| 567 |
+
"eval_samples_per_second": 105.821,
|
| 568 |
+
"eval_steps_per_second": 6.719,
|
| 569 |
+
"step": 350
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 6.67,
|
| 573 |
+
"grad_norm": 2.7061989307403564,
|
| 574 |
+
"learning_rate": 0.00013351851851851852,
|
| 575 |
+
"loss": 0.1297,
|
| 576 |
+
"step": 360
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 6.67,
|
| 580 |
+
"eval_accuracy": 0.6772486772486772,
|
| 581 |
+
"eval_loss": 1.1142699718475342,
|
| 582 |
+
"eval_runtime": 1.7368,
|
| 583 |
+
"eval_samples_per_second": 108.818,
|
| 584 |
+
"eval_steps_per_second": 6.909,
|
| 585 |
+
"step": 360
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 6.85,
|
| 589 |
+
"grad_norm": 2.478459358215332,
|
| 590 |
+
"learning_rate": 0.00013166666666666668,
|
| 591 |
+
"loss": 0.2191,
|
| 592 |
+
"step": 370
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 6.85,
|
| 596 |
+
"eval_accuracy": 0.7354497354497355,
|
| 597 |
+
"eval_loss": 0.9896882176399231,
|
| 598 |
+
"eval_runtime": 1.7802,
|
| 599 |
+
"eval_samples_per_second": 106.167,
|
| 600 |
+
"eval_steps_per_second": 6.741,
|
| 601 |
+
"step": 370
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 7.04,
|
| 605 |
+
"grad_norm": 1.6921576261520386,
|
| 606 |
+
"learning_rate": 0.0001298148148148148,
|
| 607 |
+
"loss": 0.1206,
|
| 608 |
+
"step": 380
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 7.04,
|
| 612 |
+
"eval_accuracy": 0.7407407407407407,
|
| 613 |
+
"eval_loss": 0.962655782699585,
|
| 614 |
+
"eval_runtime": 1.7667,
|
| 615 |
+
"eval_samples_per_second": 106.982,
|
| 616 |
+
"eval_steps_per_second": 6.793,
|
| 617 |
+
"step": 380
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 7.22,
|
| 621 |
+
"grad_norm": 0.8060858845710754,
|
| 622 |
+
"learning_rate": 0.00012796296296296297,
|
| 623 |
+
"loss": 0.1071,
|
| 624 |
+
"step": 390
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 7.22,
|
| 628 |
+
"eval_accuracy": 0.7513227513227513,
|
| 629 |
+
"eval_loss": 1.0495431423187256,
|
| 630 |
+
"eval_runtime": 1.7473,
|
| 631 |
+
"eval_samples_per_second": 108.168,
|
| 632 |
+
"eval_steps_per_second": 6.868,
|
| 633 |
+
"step": 390
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 7.41,
|
| 637 |
+
"grad_norm": 0.38671812415122986,
|
| 638 |
+
"learning_rate": 0.0001261111111111111,
|
| 639 |
+
"loss": 0.1102,
|
| 640 |
+
"step": 400
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 7.41,
|
| 644 |
+
"eval_accuracy": 0.7301587301587301,
|
| 645 |
+
"eval_loss": 1.0441887378692627,
|
| 646 |
+
"eval_runtime": 1.7747,
|
| 647 |
+
"eval_samples_per_second": 106.496,
|
| 648 |
+
"eval_steps_per_second": 6.762,
|
| 649 |
+
"step": 400
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 7.59,
|
| 653 |
+
"grad_norm": 1.2801034450531006,
|
| 654 |
+
"learning_rate": 0.0001242592592592593,
|
| 655 |
+
"loss": 0.1269,
|
| 656 |
+
"step": 410
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 7.59,
|
| 660 |
+
"eval_accuracy": 0.7407407407407407,
|
| 661 |
+
"eval_loss": 1.0281165838241577,
|
| 662 |
+
"eval_runtime": 1.811,
|
| 663 |
+
"eval_samples_per_second": 104.363,
|
| 664 |
+
"eval_steps_per_second": 6.626,
|
| 665 |
+
"step": 410
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 7.78,
|
| 669 |
+
"grad_norm": 0.92644864320755,
|
| 670 |
+
"learning_rate": 0.00012240740740740742,
|
| 671 |
+
"loss": 0.0694,
|
| 672 |
+
"step": 420
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 7.78,
|
| 676 |
+
"eval_accuracy": 0.7354497354497355,
|
| 677 |
+
"eval_loss": 1.0361741781234741,
|
| 678 |
+
"eval_runtime": 1.7423,
|
| 679 |
+
"eval_samples_per_second": 108.479,
|
| 680 |
+
"eval_steps_per_second": 6.888,
|
| 681 |
+
"step": 420
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 7.96,
|
| 685 |
+
"grad_norm": 0.8203582167625427,
|
| 686 |
+
"learning_rate": 0.00012055555555555555,
|
| 687 |
+
"loss": 0.0548,
|
| 688 |
+
"step": 430
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 7.96,
|
| 692 |
+
"eval_accuracy": 0.746031746031746,
|
| 693 |
+
"eval_loss": 1.071204423904419,
|
| 694 |
+
"eval_runtime": 1.7384,
|
| 695 |
+
"eval_samples_per_second": 108.723,
|
| 696 |
+
"eval_steps_per_second": 6.903,
|
| 697 |
+
"step": 430
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 8.15,
|
| 701 |
+
"grad_norm": 4.490820407867432,
|
| 702 |
+
"learning_rate": 0.00011870370370370371,
|
| 703 |
+
"loss": 0.062,
|
| 704 |
+
"step": 440
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 8.15,
|
| 708 |
+
"eval_accuracy": 0.7301587301587301,
|
| 709 |
+
"eval_loss": 1.035632610321045,
|
| 710 |
+
"eval_runtime": 1.8141,
|
| 711 |
+
"eval_samples_per_second": 104.182,
|
| 712 |
+
"eval_steps_per_second": 6.615,
|
| 713 |
+
"step": 440
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 8.33,
|
| 717 |
+
"grad_norm": 1.979749321937561,
|
| 718 |
+
"learning_rate": 0.00011685185185185186,
|
| 719 |
+
"loss": 0.0542,
|
| 720 |
+
"step": 450
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 8.33,
|
| 724 |
+
"eval_accuracy": 0.6984126984126984,
|
| 725 |
+
"eval_loss": 1.2573037147521973,
|
| 726 |
+
"eval_runtime": 1.7824,
|
| 727 |
+
"eval_samples_per_second": 106.034,
|
| 728 |
+
"eval_steps_per_second": 6.732,
|
| 729 |
+
"step": 450
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 8.52,
|
| 733 |
+
"grad_norm": 4.157647609710693,
|
| 734 |
+
"learning_rate": 0.00011499999999999999,
|
| 735 |
+
"loss": 0.0823,
|
| 736 |
+
"step": 460
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 8.52,
|
| 740 |
+
"eval_accuracy": 0.7195767195767195,
|
| 741 |
+
"eval_loss": 1.1037700176239014,
|
| 742 |
+
"eval_runtime": 1.7489,
|
| 743 |
+
"eval_samples_per_second": 108.066,
|
| 744 |
+
"eval_steps_per_second": 6.861,
|
| 745 |
+
"step": 460
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 8.7,
|
| 749 |
+
"grad_norm": 0.08767159283161163,
|
| 750 |
+
"learning_rate": 0.00011314814814814816,
|
| 751 |
+
"loss": 0.1354,
|
| 752 |
+
"step": 470
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 8.7,
|
| 756 |
+
"eval_accuracy": 0.7407407407407407,
|
| 757 |
+
"eval_loss": 1.0803223848342896,
|
| 758 |
+
"eval_runtime": 1.7889,
|
| 759 |
+
"eval_samples_per_second": 105.654,
|
| 760 |
+
"eval_steps_per_second": 6.708,
|
| 761 |
+
"step": 470
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 8.89,
|
| 765 |
+
"grad_norm": 0.6974061131477356,
|
| 766 |
+
"learning_rate": 0.0001112962962962963,
|
| 767 |
+
"loss": 0.0798,
|
| 768 |
+
"step": 480
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 8.89,
|
| 772 |
+
"eval_accuracy": 0.671957671957672,
|
| 773 |
+
"eval_loss": 1.2207469940185547,
|
| 774 |
+
"eval_runtime": 1.7456,
|
| 775 |
+
"eval_samples_per_second": 108.27,
|
| 776 |
+
"eval_steps_per_second": 6.874,
|
| 777 |
+
"step": 480
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 9.07,
|
| 781 |
+
"grad_norm": 2.0027213096618652,
|
| 782 |
+
"learning_rate": 0.00010944444444444445,
|
| 783 |
+
"loss": 0.0963,
|
| 784 |
+
"step": 490
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 9.07,
|
| 788 |
+
"eval_accuracy": 0.656084656084656,
|
| 789 |
+
"eval_loss": 1.337466835975647,
|
| 790 |
+
"eval_runtime": 1.7654,
|
| 791 |
+
"eval_samples_per_second": 107.06,
|
| 792 |
+
"eval_steps_per_second": 6.797,
|
| 793 |
+
"step": 490
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 9.26,
|
| 797 |
+
"grad_norm": 0.14471650123596191,
|
| 798 |
+
"learning_rate": 0.0001075925925925926,
|
| 799 |
+
"loss": 0.0557,
|
| 800 |
+
"step": 500
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 9.26,
|
| 804 |
+
"eval_accuracy": 0.6984126984126984,
|
| 805 |
+
"eval_loss": 1.2044044733047485,
|
| 806 |
+
"eval_runtime": 1.9948,
|
| 807 |
+
"eval_samples_per_second": 94.745,
|
| 808 |
+
"eval_steps_per_second": 6.016,
|
| 809 |
+
"step": 500
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 9.44,
|
| 813 |
+
"grad_norm": 0.07393530756235123,
|
| 814 |
+
"learning_rate": 0.00010574074074074075,
|
| 815 |
+
"loss": 0.0491,
|
| 816 |
+
"step": 510
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 9.44,
|
| 820 |
+
"eval_accuracy": 0.7248677248677249,
|
| 821 |
+
"eval_loss": 1.18802809715271,
|
| 822 |
+
"eval_runtime": 1.8204,
|
| 823 |
+
"eval_samples_per_second": 103.822,
|
| 824 |
+
"eval_steps_per_second": 6.592,
|
| 825 |
+
"step": 510
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 9.63,
|
| 829 |
+
"grad_norm": 0.12744389474391937,
|
| 830 |
+
"learning_rate": 0.0001038888888888889,
|
| 831 |
+
"loss": 0.0502,
|
| 832 |
+
"step": 520
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 9.63,
|
| 836 |
+
"eval_accuracy": 0.746031746031746,
|
| 837 |
+
"eval_loss": 1.098527193069458,
|
| 838 |
+
"eval_runtime": 1.7601,
|
| 839 |
+
"eval_samples_per_second": 107.378,
|
| 840 |
+
"eval_steps_per_second": 6.818,
|
| 841 |
+
"step": 520
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 9.81,
|
| 845 |
+
"grad_norm": 0.07471567392349243,
|
| 846 |
+
"learning_rate": 0.00010203703703703704,
|
| 847 |
+
"loss": 0.0396,
|
| 848 |
+
"step": 530
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 9.81,
|
| 852 |
+
"eval_accuracy": 0.708994708994709,
|
| 853 |
+
"eval_loss": 1.214396595954895,
|
| 854 |
+
"eval_runtime": 1.7884,
|
| 855 |
+
"eval_samples_per_second": 105.68,
|
| 856 |
+
"eval_steps_per_second": 6.71,
|
| 857 |
+
"step": 530
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 10.0,
|
| 861 |
+
"grad_norm": 0.16710619628429413,
|
| 862 |
+
"learning_rate": 0.00010018518518518518,
|
| 863 |
+
"loss": 0.0717,
|
| 864 |
+
"step": 540
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 10.0,
|
| 868 |
+
"eval_accuracy": 0.7037037037037037,
|
| 869 |
+
"eval_loss": 1.2163357734680176,
|
| 870 |
+
"eval_runtime": 1.7401,
|
| 871 |
+
"eval_samples_per_second": 108.615,
|
| 872 |
+
"eval_steps_per_second": 6.896,
|
| 873 |
+
"step": 540
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 10.19,
|
| 877 |
+
"grad_norm": 0.07553374022245407,
|
| 878 |
+
"learning_rate": 9.833333333333333e-05,
|
| 879 |
+
"loss": 0.0279,
|
| 880 |
+
"step": 550
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 10.19,
|
| 884 |
+
"eval_accuracy": 0.7142857142857143,
|
| 885 |
+
"eval_loss": 1.119241714477539,
|
| 886 |
+
"eval_runtime": 1.766,
|
| 887 |
+
"eval_samples_per_second": 107.023,
|
| 888 |
+
"eval_steps_per_second": 6.795,
|
| 889 |
+
"step": 550
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 10.37,
|
| 893 |
+
"grad_norm": 0.07353632897138596,
|
| 894 |
+
"learning_rate": 9.648148148148149e-05,
|
| 895 |
+
"loss": 0.0329,
|
| 896 |
+
"step": 560
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 10.37,
|
| 900 |
+
"eval_accuracy": 0.7354497354497355,
|
| 901 |
+
"eval_loss": 1.1961112022399902,
|
| 902 |
+
"eval_runtime": 1.8216,
|
| 903 |
+
"eval_samples_per_second": 103.758,
|
| 904 |
+
"eval_steps_per_second": 6.588,
|
| 905 |
+
"step": 560
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 10.56,
|
| 909 |
+
"grad_norm": 0.5441647171974182,
|
| 910 |
+
"learning_rate": 9.462962962962963e-05,
|
| 911 |
+
"loss": 0.028,
|
| 912 |
+
"step": 570
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 10.56,
|
| 916 |
+
"eval_accuracy": 0.6984126984126984,
|
| 917 |
+
"eval_loss": 1.1282387971878052,
|
| 918 |
+
"eval_runtime": 1.7883,
|
| 919 |
+
"eval_samples_per_second": 105.689,
|
| 920 |
+
"eval_steps_per_second": 6.71,
|
| 921 |
+
"step": 570
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 10.74,
|
| 925 |
+
"grad_norm": 0.07243653386831284,
|
| 926 |
+
"learning_rate": 9.277777777777778e-05,
|
| 927 |
+
"loss": 0.0373,
|
| 928 |
+
"step": 580
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 10.74,
|
| 932 |
+
"eval_accuracy": 0.7195767195767195,
|
| 933 |
+
"eval_loss": 1.0716224908828735,
|
| 934 |
+
"eval_runtime": 1.736,
|
| 935 |
+
"eval_samples_per_second": 108.873,
|
| 936 |
+
"eval_steps_per_second": 6.913,
|
| 937 |
+
"step": 580
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 10.93,
|
| 941 |
+
"grad_norm": 0.04851379618048668,
|
| 942 |
+
"learning_rate": 9.092592592592593e-05,
|
| 943 |
+
"loss": 0.0368,
|
| 944 |
+
"step": 590
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 10.93,
|
| 948 |
+
"eval_accuracy": 0.7142857142857143,
|
| 949 |
+
"eval_loss": 1.1750774383544922,
|
| 950 |
+
"eval_runtime": 1.7848,
|
| 951 |
+
"eval_samples_per_second": 105.895,
|
| 952 |
+
"eval_steps_per_second": 6.723,
|
| 953 |
+
"step": 590
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 11.11,
|
| 957 |
+
"grad_norm": 0.05160636082291603,
|
| 958 |
+
"learning_rate": 8.907407407407407e-05,
|
| 959 |
+
"loss": 0.0485,
|
| 960 |
+
"step": 600
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 11.11,
|
| 964 |
+
"eval_accuracy": 0.7354497354497355,
|
| 965 |
+
"eval_loss": 1.0984432697296143,
|
| 966 |
+
"eval_runtime": 1.7772,
|
| 967 |
+
"eval_samples_per_second": 106.345,
|
| 968 |
+
"eval_steps_per_second": 6.752,
|
| 969 |
+
"step": 600
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 11.3,
|
| 973 |
+
"grad_norm": 0.054380565881729126,
|
| 974 |
+
"learning_rate": 8.722222222222223e-05,
|
| 975 |
+
"loss": 0.0234,
|
| 976 |
+
"step": 610
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 11.3,
|
| 980 |
+
"eval_accuracy": 0.7619047619047619,
|
| 981 |
+
"eval_loss": 1.0418734550476074,
|
| 982 |
+
"eval_runtime": 1.7977,
|
| 983 |
+
"eval_samples_per_second": 105.132,
|
| 984 |
+
"eval_steps_per_second": 6.675,
|
| 985 |
+
"step": 610
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 11.48,
|
| 989 |
+
"grad_norm": 0.32195061445236206,
|
| 990 |
+
"learning_rate": 8.537037037037038e-05,
|
| 991 |
+
"loss": 0.028,
|
| 992 |
+
"step": 620
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 11.48,
|
| 996 |
+
"eval_accuracy": 0.7566137566137566,
|
| 997 |
+
"eval_loss": 1.0536975860595703,
|
| 998 |
+
"eval_runtime": 1.7586,
|
| 999 |
+
"eval_samples_per_second": 107.47,
|
| 1000 |
+
"eval_steps_per_second": 6.823,
|
| 1001 |
+
"step": 620
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 11.67,
|
| 1005 |
+
"grad_norm": 1.8460614681243896,
|
| 1006 |
+
"learning_rate": 8.351851851851852e-05,
|
| 1007 |
+
"loss": 0.0237,
|
| 1008 |
+
"step": 630
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 11.67,
|
| 1012 |
+
"eval_accuracy": 0.746031746031746,
|
| 1013 |
+
"eval_loss": 1.0571786165237427,
|
| 1014 |
+
"eval_runtime": 1.7901,
|
| 1015 |
+
"eval_samples_per_second": 105.578,
|
| 1016 |
+
"eval_steps_per_second": 6.703,
|
| 1017 |
+
"step": 630
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 11.85,
|
| 1021 |
+
"grad_norm": 1.7614848613739014,
|
| 1022 |
+
"learning_rate": 8.166666666666667e-05,
|
| 1023 |
+
"loss": 0.0198,
|
| 1024 |
+
"step": 640
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 11.85,
|
| 1028 |
+
"eval_accuracy": 0.746031746031746,
|
| 1029 |
+
"eval_loss": 1.0192136764526367,
|
| 1030 |
+
"eval_runtime": 1.7683,
|
| 1031 |
+
"eval_samples_per_second": 106.885,
|
| 1032 |
+
"eval_steps_per_second": 6.786,
|
| 1033 |
+
"step": 640
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 12.04,
|
| 1037 |
+
"grad_norm": 0.22871683537960052,
|
| 1038 |
+
"learning_rate": 7.981481481481481e-05,
|
| 1039 |
+
"loss": 0.02,
|
| 1040 |
+
"step": 650
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 12.04,
|
| 1044 |
+
"eval_accuracy": 0.7195767195767195,
|
| 1045 |
+
"eval_loss": 1.244175672531128,
|
| 1046 |
+
"eval_runtime": 1.8603,
|
| 1047 |
+
"eval_samples_per_second": 101.595,
|
| 1048 |
+
"eval_steps_per_second": 6.45,
|
| 1049 |
+
"step": 650
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 12.22,
|
| 1053 |
+
"grad_norm": 0.03752712532877922,
|
| 1054 |
+
"learning_rate": 7.796296296296297e-05,
|
| 1055 |
+
"loss": 0.0216,
|
| 1056 |
+
"step": 660
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 12.22,
|
| 1060 |
+
"eval_accuracy": 0.7407407407407407,
|
| 1061 |
+
"eval_loss": 1.1395213603973389,
|
| 1062 |
+
"eval_runtime": 1.7992,
|
| 1063 |
+
"eval_samples_per_second": 105.048,
|
| 1064 |
+
"eval_steps_per_second": 6.67,
|
| 1065 |
+
"step": 660
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 12.41,
|
| 1069 |
+
"grad_norm": 0.09251418709754944,
|
| 1070 |
+
"learning_rate": 7.61111111111111e-05,
|
| 1071 |
+
"loss": 0.0309,
|
| 1072 |
+
"step": 670
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 12.41,
|
| 1076 |
+
"eval_accuracy": 0.7354497354497355,
|
| 1077 |
+
"eval_loss": 1.1767151355743408,
|
| 1078 |
+
"eval_runtime": 1.8204,
|
| 1079 |
+
"eval_samples_per_second": 103.823,
|
| 1080 |
+
"eval_steps_per_second": 6.592,
|
| 1081 |
+
"step": 670
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 12.59,
|
| 1085 |
+
"grad_norm": 0.03858701139688492,
|
| 1086 |
+
"learning_rate": 7.425925925925927e-05,
|
| 1087 |
+
"loss": 0.0315,
|
| 1088 |
+
"step": 680
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 12.59,
|
| 1092 |
+
"eval_accuracy": 0.7248677248677249,
|
| 1093 |
+
"eval_loss": 1.1881897449493408,
|
| 1094 |
+
"eval_runtime": 1.7853,
|
| 1095 |
+
"eval_samples_per_second": 105.862,
|
| 1096 |
+
"eval_steps_per_second": 6.721,
|
| 1097 |
+
"step": 680
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 12.78,
|
| 1101 |
+
"grad_norm": 0.2986956536769867,
|
| 1102 |
+
"learning_rate": 7.240740740740741e-05,
|
| 1103 |
+
"loss": 0.017,
|
| 1104 |
+
"step": 690
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 12.78,
|
| 1108 |
+
"eval_accuracy": 0.7354497354497355,
|
| 1109 |
+
"eval_loss": 1.1652072668075562,
|
| 1110 |
+
"eval_runtime": 1.8006,
|
| 1111 |
+
"eval_samples_per_second": 104.965,
|
| 1112 |
+
"eval_steps_per_second": 6.664,
|
| 1113 |
+
"step": 690
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 12.96,
|
| 1117 |
+
"grad_norm": 0.23789283633232117,
|
| 1118 |
+
"learning_rate": 7.055555555555556e-05,
|
| 1119 |
+
"loss": 0.02,
|
| 1120 |
+
"step": 700
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 12.96,
|
| 1124 |
+
"eval_accuracy": 0.7619047619047619,
|
| 1125 |
+
"eval_loss": 1.1011323928833008,
|
| 1126 |
+
"eval_runtime": 1.7393,
|
| 1127 |
+
"eval_samples_per_second": 108.665,
|
| 1128 |
+
"eval_steps_per_second": 6.899,
|
| 1129 |
+
"step": 700
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 13.15,
|
| 1133 |
+
"grad_norm": 0.0361974723637104,
|
| 1134 |
+
"learning_rate": 6.87037037037037e-05,
|
| 1135 |
+
"loss": 0.0174,
|
| 1136 |
+
"step": 710
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 13.15,
|
| 1140 |
+
"eval_accuracy": 0.7354497354497355,
|
| 1141 |
+
"eval_loss": 1.092558741569519,
|
| 1142 |
+
"eval_runtime": 1.8005,
|
| 1143 |
+
"eval_samples_per_second": 104.97,
|
| 1144 |
+
"eval_steps_per_second": 6.665,
|
| 1145 |
+
"step": 710
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 13.33,
|
| 1149 |
+
"grad_norm": 0.04739515110850334,
|
| 1150 |
+
"learning_rate": 6.685185185185185e-05,
|
| 1151 |
+
"loss": 0.012,
|
| 1152 |
+
"step": 720
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 13.33,
|
| 1156 |
+
"eval_accuracy": 0.746031746031746,
|
| 1157 |
+
"eval_loss": 1.0852241516113281,
|
| 1158 |
+
"eval_runtime": 1.787,
|
| 1159 |
+
"eval_samples_per_second": 105.766,
|
| 1160 |
+
"eval_steps_per_second": 6.715,
|
| 1161 |
+
"step": 720
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 13.52,
|
| 1165 |
+
"grad_norm": 0.035977743566036224,
|
| 1166 |
+
"learning_rate": 6.500000000000001e-05,
|
| 1167 |
+
"loss": 0.0296,
|
| 1168 |
+
"step": 730
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 13.52,
|
| 1172 |
+
"eval_accuracy": 0.7513227513227513,
|
| 1173 |
+
"eval_loss": 1.0534002780914307,
|
| 1174 |
+
"eval_runtime": 1.7706,
|
| 1175 |
+
"eval_samples_per_second": 106.746,
|
| 1176 |
+
"eval_steps_per_second": 6.778,
|
| 1177 |
+
"step": 730
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"epoch": 13.7,
|
| 1181 |
+
"grad_norm": 0.3354228436946869,
|
| 1182 |
+
"learning_rate": 6.314814814814815e-05,
|
| 1183 |
+
"loss": 0.0142,
|
| 1184 |
+
"step": 740
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 13.7,
|
| 1188 |
+
"eval_accuracy": 0.746031746031746,
|
| 1189 |
+
"eval_loss": 1.0607830286026,
|
| 1190 |
+
"eval_runtime": 1.8039,
|
| 1191 |
+
"eval_samples_per_second": 104.775,
|
| 1192 |
+
"eval_steps_per_second": 6.652,
|
| 1193 |
+
"step": 740
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 13.89,
|
| 1197 |
+
"grad_norm": 0.031177503988146782,
|
| 1198 |
+
"learning_rate": 6.12962962962963e-05,
|
| 1199 |
+
"loss": 0.0199,
|
| 1200 |
+
"step": 750
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 13.89,
|
| 1204 |
+
"eval_accuracy": 0.746031746031746,
|
| 1205 |
+
"eval_loss": 1.0850036144256592,
|
| 1206 |
+
"eval_runtime": 1.7472,
|
| 1207 |
+
"eval_samples_per_second": 108.174,
|
| 1208 |
+
"eval_steps_per_second": 6.868,
|
| 1209 |
+
"step": 750
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 14.07,
|
| 1213 |
+
"grad_norm": 0.2141834944486618,
|
| 1214 |
+
"learning_rate": 5.9444444444444445e-05,
|
| 1215 |
+
"loss": 0.0169,
|
| 1216 |
+
"step": 760
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 14.07,
|
| 1220 |
+
"eval_accuracy": 0.7566137566137566,
|
| 1221 |
+
"eval_loss": 1.0736693143844604,
|
| 1222 |
+
"eval_runtime": 1.7821,
|
| 1223 |
+
"eval_samples_per_second": 106.054,
|
| 1224 |
+
"eval_steps_per_second": 6.734,
|
| 1225 |
+
"step": 760
|
| 1226 |
+
},
|
| 1227 |
+
{
|
| 1228 |
+
"epoch": 14.26,
|
| 1229 |
+
"grad_norm": 0.028399189934134483,
|
| 1230 |
+
"learning_rate": 5.75925925925926e-05,
|
| 1231 |
+
"loss": 0.0139,
|
| 1232 |
+
"step": 770
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 14.26,
|
| 1236 |
+
"eval_accuracy": 0.7566137566137566,
|
| 1237 |
+
"eval_loss": 1.0717233419418335,
|
| 1238 |
+
"eval_runtime": 1.8135,
|
| 1239 |
+
"eval_samples_per_second": 104.221,
|
| 1240 |
+
"eval_steps_per_second": 6.617,
|
| 1241 |
+
"step": 770
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 14.44,
|
| 1245 |
+
"grad_norm": 0.03289506584405899,
|
| 1246 |
+
"learning_rate": 5.574074074074075e-05,
|
| 1247 |
+
"loss": 0.0173,
|
| 1248 |
+
"step": 780
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 14.44,
|
| 1252 |
+
"eval_accuracy": 0.7566137566137566,
|
| 1253 |
+
"eval_loss": 1.0707134008407593,
|
| 1254 |
+
"eval_runtime": 1.7856,
|
| 1255 |
+
"eval_samples_per_second": 105.849,
|
| 1256 |
+
"eval_steps_per_second": 6.721,
|
| 1257 |
+
"step": 780
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 14.63,
|
| 1261 |
+
"grad_norm": 0.032911308109760284,
|
| 1262 |
+
"learning_rate": 5.388888888888889e-05,
|
| 1263 |
+
"loss": 0.0101,
|
| 1264 |
+
"step": 790
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 14.63,
|
| 1268 |
+
"eval_accuracy": 0.7566137566137566,
|
| 1269 |
+
"eval_loss": 1.070402979850769,
|
| 1270 |
+
"eval_runtime": 1.7933,
|
| 1271 |
+
"eval_samples_per_second": 105.391,
|
| 1272 |
+
"eval_steps_per_second": 6.691,
|
| 1273 |
+
"step": 790
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 14.81,
|
| 1277 |
+
"grad_norm": 0.43361806869506836,
|
| 1278 |
+
"learning_rate": 5.203703703703704e-05,
|
| 1279 |
+
"loss": 0.0286,
|
| 1280 |
+
"step": 800
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 14.81,
|
| 1284 |
+
"eval_accuracy": 0.7671957671957672,
|
| 1285 |
+
"eval_loss": 1.0845017433166504,
|
| 1286 |
+
"eval_runtime": 1.7994,
|
| 1287 |
+
"eval_samples_per_second": 105.033,
|
| 1288 |
+
"eval_steps_per_second": 6.669,
|
| 1289 |
+
"step": 800
|
| 1290 |
+
},
|
| 1291 |
+
{
|
| 1292 |
+
"epoch": 15.0,
|
| 1293 |
+
"grad_norm": 0.05939367786049843,
|
| 1294 |
+
"learning_rate": 5.018518518518519e-05,
|
| 1295 |
+
"loss": 0.0135,
|
| 1296 |
+
"step": 810
|
| 1297 |
+
},
|
| 1298 |
+
{
|
| 1299 |
+
"epoch": 15.0,
|
| 1300 |
+
"eval_accuracy": 0.7513227513227513,
|
| 1301 |
+
"eval_loss": 1.0972745418548584,
|
| 1302 |
+
"eval_runtime": 1.7785,
|
| 1303 |
+
"eval_samples_per_second": 106.271,
|
| 1304 |
+
"eval_steps_per_second": 6.747,
|
| 1305 |
+
"step": 810
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 15.19,
|
| 1309 |
+
"grad_norm": 0.030746394768357277,
|
| 1310 |
+
"learning_rate": 4.8333333333333334e-05,
|
| 1311 |
+
"loss": 0.0129,
|
| 1312 |
+
"step": 820
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 15.19,
|
| 1316 |
+
"eval_accuracy": 0.7513227513227513,
|
| 1317 |
+
"eval_loss": 1.0909744501113892,
|
| 1318 |
+
"eval_runtime": 1.7304,
|
| 1319 |
+
"eval_samples_per_second": 109.222,
|
| 1320 |
+
"eval_steps_per_second": 6.935,
|
| 1321 |
+
"step": 820
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 15.37,
|
| 1325 |
+
"grad_norm": 0.026390748098492622,
|
| 1326 |
+
"learning_rate": 4.648148148148148e-05,
|
| 1327 |
+
"loss": 0.0117,
|
| 1328 |
+
"step": 830
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 15.37,
|
| 1332 |
+
"eval_accuracy": 0.7671957671957672,
|
| 1333 |
+
"eval_loss": 1.0890551805496216,
|
| 1334 |
+
"eval_runtime": 1.8164,
|
| 1335 |
+
"eval_samples_per_second": 104.051,
|
| 1336 |
+
"eval_steps_per_second": 6.606,
|
| 1337 |
+
"step": 830
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 15.56,
|
| 1341 |
+
"grad_norm": 0.028341053053736687,
|
| 1342 |
+
"learning_rate": 4.462962962962963e-05,
|
| 1343 |
+
"loss": 0.014,
|
| 1344 |
+
"step": 840
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 15.56,
|
| 1348 |
+
"eval_accuracy": 0.7566137566137566,
|
| 1349 |
+
"eval_loss": 1.0884122848510742,
|
| 1350 |
+
"eval_runtime": 1.8336,
|
| 1351 |
+
"eval_samples_per_second": 103.079,
|
| 1352 |
+
"eval_steps_per_second": 6.545,
|
| 1353 |
+
"step": 840
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 15.74,
|
| 1357 |
+
"grad_norm": 0.027172435075044632,
|
| 1358 |
+
"learning_rate": 4.277777777777778e-05,
|
| 1359 |
+
"loss": 0.0093,
|
| 1360 |
+
"step": 850
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 15.74,
|
| 1364 |
+
"eval_accuracy": 0.7513227513227513,
|
| 1365 |
+
"eval_loss": 1.0879539251327515,
|
| 1366 |
+
"eval_runtime": 1.7368,
|
| 1367 |
+
"eval_samples_per_second": 108.818,
|
| 1368 |
+
"eval_steps_per_second": 6.909,
|
| 1369 |
+
"step": 850
|
| 1370 |
+
},
|
| 1371 |
+
{
|
| 1372 |
+
"epoch": 15.93,
|
| 1373 |
+
"grad_norm": 0.4558853209018707,
|
| 1374 |
+
"learning_rate": 4.092592592592593e-05,
|
| 1375 |
+
"loss": 0.0264,
|
| 1376 |
+
"step": 860
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 15.93,
|
| 1380 |
+
"eval_accuracy": 0.7566137566137566,
|
| 1381 |
+
"eval_loss": 1.0861279964447021,
|
| 1382 |
+
"eval_runtime": 1.8295,
|
| 1383 |
+
"eval_samples_per_second": 103.306,
|
| 1384 |
+
"eval_steps_per_second": 6.559,
|
| 1385 |
+
"step": 860
|
| 1386 |
+
},
|
| 1387 |
+
{
|
| 1388 |
+
"epoch": 16.11,
|
| 1389 |
+
"grad_norm": 0.023086287081241608,
|
| 1390 |
+
"learning_rate": 3.9074074074074076e-05,
|
| 1391 |
+
"loss": 0.0117,
|
| 1392 |
+
"step": 870
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"epoch": 16.11,
|
| 1396 |
+
"eval_accuracy": 0.7513227513227513,
|
| 1397 |
+
"eval_loss": 1.0812128782272339,
|
| 1398 |
+
"eval_runtime": 1.783,
|
| 1399 |
+
"eval_samples_per_second": 106.0,
|
| 1400 |
+
"eval_steps_per_second": 6.73,
|
| 1401 |
+
"step": 870
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"epoch": 16.3,
|
| 1405 |
+
"grad_norm": 0.16555258631706238,
|
| 1406 |
+
"learning_rate": 3.722222222222222e-05,
|
| 1407 |
+
"loss": 0.0131,
|
| 1408 |
+
"step": 880
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 16.3,
|
| 1412 |
+
"eval_accuracy": 0.7513227513227513,
|
| 1413 |
+
"eval_loss": 1.084083080291748,
|
| 1414 |
+
"eval_runtime": 1.7979,
|
| 1415 |
+
"eval_samples_per_second": 105.125,
|
| 1416 |
+
"eval_steps_per_second": 6.675,
|
| 1417 |
+
"step": 880
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 16.48,
|
| 1421 |
+
"grad_norm": 0.1985342651605606,
|
| 1422 |
+
"learning_rate": 3.537037037037037e-05,
|
| 1423 |
+
"loss": 0.0107,
|
| 1424 |
+
"step": 890
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 16.48,
|
| 1428 |
+
"eval_accuracy": 0.7513227513227513,
|
| 1429 |
+
"eval_loss": 1.0908081531524658,
|
| 1430 |
+
"eval_runtime": 1.8371,
|
| 1431 |
+
"eval_samples_per_second": 102.877,
|
| 1432 |
+
"eval_steps_per_second": 6.532,
|
| 1433 |
+
"step": 890
|
| 1434 |
+
},
|
| 1435 |
+
{
|
| 1436 |
+
"epoch": 16.67,
|
| 1437 |
+
"grad_norm": 0.023619532585144043,
|
| 1438 |
+
"learning_rate": 3.351851851851852e-05,
|
| 1439 |
+
"loss": 0.0253,
|
| 1440 |
+
"step": 900
|
| 1441 |
+
},
|
| 1442 |
+
{
|
| 1443 |
+
"epoch": 16.67,
|
| 1444 |
+
"eval_accuracy": 0.7566137566137566,
|
| 1445 |
+
"eval_loss": 1.0818437337875366,
|
| 1446 |
+
"eval_runtime": 1.8128,
|
| 1447 |
+
"eval_samples_per_second": 104.258,
|
| 1448 |
+
"eval_steps_per_second": 6.62,
|
| 1449 |
+
"step": 900
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 16.85,
|
| 1453 |
+
"grad_norm": 0.031866107136011124,
|
| 1454 |
+
"learning_rate": 3.1666666666666666e-05,
|
| 1455 |
+
"loss": 0.0113,
|
| 1456 |
+
"step": 910
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 16.85,
|
| 1460 |
+
"eval_accuracy": 0.7671957671957672,
|
| 1461 |
+
"eval_loss": 1.0804176330566406,
|
| 1462 |
+
"eval_runtime": 1.7557,
|
| 1463 |
+
"eval_samples_per_second": 107.647,
|
| 1464 |
+
"eval_steps_per_second": 6.835,
|
| 1465 |
+
"step": 910
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 17.04,
|
| 1469 |
+
"grad_norm": 0.027054764330387115,
|
| 1470 |
+
"learning_rate": 2.981481481481482e-05,
|
| 1471 |
+
"loss": 0.0117,
|
| 1472 |
+
"step": 920
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 17.04,
|
| 1476 |
+
"eval_accuracy": 0.7671957671957672,
|
| 1477 |
+
"eval_loss": 1.0813896656036377,
|
| 1478 |
+
"eval_runtime": 1.8358,
|
| 1479 |
+
"eval_samples_per_second": 102.952,
|
| 1480 |
+
"eval_steps_per_second": 6.537,
|
| 1481 |
+
"step": 920
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 17.22,
|
| 1485 |
+
"grad_norm": 0.025050414726138115,
|
| 1486 |
+
"learning_rate": 2.7962962962962965e-05,
|
| 1487 |
+
"loss": 0.0158,
|
| 1488 |
+
"step": 930
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 17.22,
|
| 1492 |
+
"eval_accuracy": 0.7566137566137566,
|
| 1493 |
+
"eval_loss": 1.0813225507736206,
|
| 1494 |
+
"eval_runtime": 1.7643,
|
| 1495 |
+
"eval_samples_per_second": 107.126,
|
| 1496 |
+
"eval_steps_per_second": 6.802,
|
| 1497 |
+
"step": 930
|
| 1498 |
+
},
|
| 1499 |
+
{
|
| 1500 |
+
"epoch": 17.41,
|
| 1501 |
+
"grad_norm": 0.024830004200339317,
|
| 1502 |
+
"learning_rate": 2.6111111111111114e-05,
|
| 1503 |
+
"loss": 0.011,
|
| 1504 |
+
"step": 940
|
| 1505 |
+
},
|
| 1506 |
+
{
|
| 1507 |
+
"epoch": 17.41,
|
| 1508 |
+
"eval_accuracy": 0.7671957671957672,
|
| 1509 |
+
"eval_loss": 1.080676794052124,
|
| 1510 |
+
"eval_runtime": 1.759,
|
| 1511 |
+
"eval_samples_per_second": 107.45,
|
| 1512 |
+
"eval_steps_per_second": 6.822,
|
| 1513 |
+
"step": 940
|
| 1514 |
+
},
|
| 1515 |
+
{
|
| 1516 |
+
"epoch": 17.59,
|
| 1517 |
+
"grad_norm": 0.024760620668530464,
|
| 1518 |
+
"learning_rate": 2.425925925925926e-05,
|
| 1519 |
+
"loss": 0.0137,
|
| 1520 |
+
"step": 950
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 17.59,
|
| 1524 |
+
"eval_accuracy": 0.7671957671957672,
|
| 1525 |
+
"eval_loss": 1.0803221464157104,
|
| 1526 |
+
"eval_runtime": 1.7971,
|
| 1527 |
+
"eval_samples_per_second": 105.168,
|
| 1528 |
+
"eval_steps_per_second": 6.677,
|
| 1529 |
+
"step": 950
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 17.78,
|
| 1533 |
+
"grad_norm": 0.025229470804333687,
|
| 1534 |
+
"learning_rate": 2.240740740740741e-05,
|
| 1535 |
+
"loss": 0.0112,
|
| 1536 |
+
"step": 960
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 17.78,
|
| 1540 |
+
"eval_accuracy": 0.7619047619047619,
|
| 1541 |
+
"eval_loss": 1.0807117223739624,
|
| 1542 |
+
"eval_runtime": 1.7675,
|
| 1543 |
+
"eval_samples_per_second": 106.93,
|
| 1544 |
+
"eval_steps_per_second": 6.789,
|
| 1545 |
+
"step": 960
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 17.96,
|
| 1549 |
+
"grad_norm": 0.02313585951924324,
|
| 1550 |
+
"learning_rate": 2.0555555555555555e-05,
|
| 1551 |
+
"loss": 0.0172,
|
| 1552 |
+
"step": 970
|
| 1553 |
+
},
|
| 1554 |
+
{
|
| 1555 |
+
"epoch": 17.96,
|
| 1556 |
+
"eval_accuracy": 0.7566137566137566,
|
| 1557 |
+
"eval_loss": 1.0821946859359741,
|
| 1558 |
+
"eval_runtime": 1.8179,
|
| 1559 |
+
"eval_samples_per_second": 103.964,
|
| 1560 |
+
"eval_steps_per_second": 6.601,
|
| 1561 |
+
"step": 970
|
| 1562 |
+
},
|
| 1563 |
+
{
|
| 1564 |
+
"epoch": 18.15,
|
| 1565 |
+
"grad_norm": 0.024956317618489265,
|
| 1566 |
+
"learning_rate": 1.8703703703703704e-05,
|
| 1567 |
+
"loss": 0.0132,
|
| 1568 |
+
"step": 980
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 18.15,
|
| 1572 |
+
"eval_accuracy": 0.7619047619047619,
|
| 1573 |
+
"eval_loss": 1.0860090255737305,
|
| 1574 |
+
"eval_runtime": 1.7729,
|
| 1575 |
+
"eval_samples_per_second": 106.607,
|
| 1576 |
+
"eval_steps_per_second": 6.769,
|
| 1577 |
+
"step": 980
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 18.33,
|
| 1581 |
+
"grad_norm": 0.02182234823703766,
|
| 1582 |
+
"learning_rate": 1.6851851851851853e-05,
|
| 1583 |
+
"loss": 0.0127,
|
| 1584 |
+
"step": 990
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 18.33,
|
| 1588 |
+
"eval_accuracy": 0.7619047619047619,
|
| 1589 |
+
"eval_loss": 1.0875723361968994,
|
| 1590 |
+
"eval_runtime": 1.7863,
|
| 1591 |
+
"eval_samples_per_second": 105.804,
|
| 1592 |
+
"eval_steps_per_second": 6.718,
|
| 1593 |
+
"step": 990
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 18.52,
|
| 1597 |
+
"grad_norm": 0.024420464411377907,
|
| 1598 |
+
"learning_rate": 1.5e-05,
|
| 1599 |
+
"loss": 0.0152,
|
| 1600 |
+
"step": 1000
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 18.52,
|
| 1604 |
+
"eval_accuracy": 0.7619047619047619,
|
| 1605 |
+
"eval_loss": 1.0873754024505615,
|
| 1606 |
+
"eval_runtime": 1.7723,
|
| 1607 |
+
"eval_samples_per_second": 106.644,
|
| 1608 |
+
"eval_steps_per_second": 6.771,
|
| 1609 |
+
"step": 1000
|
| 1610 |
+
},
|
| 1611 |
+
{
|
| 1612 |
+
"epoch": 18.7,
|
| 1613 |
+
"grad_norm": 0.18311668932437897,
|
| 1614 |
+
"learning_rate": 1.3148148148148148e-05,
|
| 1615 |
+
"loss": 0.0096,
|
| 1616 |
+
"step": 1010
|
| 1617 |
+
},
|
| 1618 |
+
{
|
| 1619 |
+
"epoch": 18.7,
|
| 1620 |
+
"eval_accuracy": 0.7619047619047619,
|
| 1621 |
+
"eval_loss": 1.088024377822876,
|
| 1622 |
+
"eval_runtime": 1.8979,
|
| 1623 |
+
"eval_samples_per_second": 99.583,
|
| 1624 |
+
"eval_steps_per_second": 6.323,
|
| 1625 |
+
"step": 1010
|
| 1626 |
+
},
|
| 1627 |
+
{
|
| 1628 |
+
"epoch": 18.89,
|
| 1629 |
+
"grad_norm": 0.023139068856835365,
|
| 1630 |
+
"learning_rate": 1.1296296296296297e-05,
|
| 1631 |
+
"loss": 0.0107,
|
| 1632 |
+
"step": 1020
|
| 1633 |
+
},
|
| 1634 |
+
{
|
| 1635 |
+
"epoch": 18.89,
|
| 1636 |
+
"eval_accuracy": 0.7619047619047619,
|
| 1637 |
+
"eval_loss": 1.08987557888031,
|
| 1638 |
+
"eval_runtime": 1.8132,
|
| 1639 |
+
"eval_samples_per_second": 104.237,
|
| 1640 |
+
"eval_steps_per_second": 6.618,
|
| 1641 |
+
"step": 1020
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 19.07,
|
| 1645 |
+
"grad_norm": 0.024323537945747375,
|
| 1646 |
+
"learning_rate": 9.444444444444445e-06,
|
| 1647 |
+
"loss": 0.0124,
|
| 1648 |
+
"step": 1030
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 19.07,
|
| 1652 |
+
"eval_accuracy": 0.7619047619047619,
|
| 1653 |
+
"eval_loss": 1.0899451971054077,
|
| 1654 |
+
"eval_runtime": 1.7841,
|
| 1655 |
+
"eval_samples_per_second": 105.934,
|
| 1656 |
+
"eval_steps_per_second": 6.726,
|
| 1657 |
+
"step": 1030
|
| 1658 |
+
},
|
| 1659 |
+
{
|
| 1660 |
+
"epoch": 19.26,
|
| 1661 |
+
"grad_norm": 0.20473988354206085,
|
| 1662 |
+
"learning_rate": 7.592592592592593e-06,
|
| 1663 |
+
"loss": 0.0187,
|
| 1664 |
+
"step": 1040
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"epoch": 19.26,
|
| 1668 |
+
"eval_accuracy": 0.7619047619047619,
|
| 1669 |
+
"eval_loss": 1.0915277004241943,
|
| 1670 |
+
"eval_runtime": 1.7828,
|
| 1671 |
+
"eval_samples_per_second": 106.015,
|
| 1672 |
+
"eval_steps_per_second": 6.731,
|
| 1673 |
+
"step": 1040
|
| 1674 |
+
},
|
| 1675 |
+
{
|
| 1676 |
+
"epoch": 19.44,
|
| 1677 |
+
"grad_norm": 0.021954894065856934,
|
| 1678 |
+
"learning_rate": 5.740740740740741e-06,
|
| 1679 |
+
"loss": 0.0159,
|
| 1680 |
+
"step": 1050
|
| 1681 |
+
},
|
| 1682 |
+
{
|
| 1683 |
+
"epoch": 19.44,
|
| 1684 |
+
"eval_accuracy": 0.7619047619047619,
|
| 1685 |
+
"eval_loss": 1.0916674137115479,
|
| 1686 |
+
"eval_runtime": 1.7554,
|
| 1687 |
+
"eval_samples_per_second": 107.665,
|
| 1688 |
+
"eval_steps_per_second": 6.836,
|
| 1689 |
+
"step": 1050
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 19.63,
|
| 1693 |
+
"grad_norm": 0.02447775937616825,
|
| 1694 |
+
"learning_rate": 3.888888888888889e-06,
|
| 1695 |
+
"loss": 0.0107,
|
| 1696 |
+
"step": 1060
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 19.63,
|
| 1700 |
+
"eval_accuracy": 0.7619047619047619,
|
| 1701 |
+
"eval_loss": 1.091030240058899,
|
| 1702 |
+
"eval_runtime": 1.7566,
|
| 1703 |
+
"eval_samples_per_second": 107.597,
|
| 1704 |
+
"eval_steps_per_second": 6.832,
|
| 1705 |
+
"step": 1060
|
| 1706 |
+
},
|
| 1707 |
+
{
|
| 1708 |
+
"epoch": 19.81,
|
| 1709 |
+
"grad_norm": 0.02190612629055977,
|
| 1710 |
+
"learning_rate": 2.0370370370370375e-06,
|
| 1711 |
+
"loss": 0.0105,
|
| 1712 |
+
"step": 1070
|
| 1713 |
+
},
|
| 1714 |
+
{
|
| 1715 |
+
"epoch": 19.81,
|
| 1716 |
+
"eval_accuracy": 0.7619047619047619,
|
| 1717 |
+
"eval_loss": 1.0911825895309448,
|
| 1718 |
+
"eval_runtime": 1.7879,
|
| 1719 |
+
"eval_samples_per_second": 105.71,
|
| 1720 |
+
"eval_steps_per_second": 6.712,
|
| 1721 |
+
"step": 1070
|
| 1722 |
+
},
|
| 1723 |
+
{
|
| 1724 |
+
"epoch": 20.0,
|
| 1725 |
+
"grad_norm": 0.0392175130546093,
|
| 1726 |
+
"learning_rate": 1.851851851851852e-07,
|
| 1727 |
+
"loss": 0.0076,
|
| 1728 |
+
"step": 1080
|
| 1729 |
+
},
|
| 1730 |
+
{
|
| 1731 |
+
"epoch": 20.0,
|
| 1732 |
+
"eval_accuracy": 0.7619047619047619,
|
| 1733 |
+
"eval_loss": 1.0909953117370605,
|
| 1734 |
+
"eval_runtime": 1.7898,
|
| 1735 |
+
"eval_samples_per_second": 105.596,
|
| 1736 |
+
"eval_steps_per_second": 6.704,
|
| 1737 |
+
"step": 1080
|
| 1738 |
+
},
|
| 1739 |
+
{
|
| 1740 |
+
"epoch": 20.0,
|
| 1741 |
+
"step": 1080,
|
| 1742 |
+
"total_flos": 2.681093741830963e+18,
|
| 1743 |
+
"train_loss": 0.29336747460895113,
|
| 1744 |
+
"train_runtime": 795.2059,
|
| 1745 |
+
"train_samples_per_second": 42.681,
|
| 1746 |
+
"train_steps_per_second": 1.358
|
| 1747 |
+
}
|
| 1748 |
+
],
|
| 1749 |
+
"logging_steps": 10,
|
| 1750 |
+
"max_steps": 1080,
|
| 1751 |
+
"num_input_tokens_seen": 0,
|
| 1752 |
+
"num_train_epochs": 20,
|
| 1753 |
+
"save_steps": 10,
|
| 1754 |
+
"total_flos": 2.681093741830963e+18,
|
| 1755 |
+
"train_batch_size": 32,
|
| 1756 |
+
"trial_name": null,
|
| 1757 |
+
"trial_params": null
|
| 1758 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5e86513f4eb3161b380f4bcf5175d22ee998a8883647c9efa0fbb1a8ca0fbded
|
| 3 |
+
size 4920
|