Upload 8 files
Browse files- config.json +69 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- preprocessor_config.json +27 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +875 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,69 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "facebook/dinov2-large",
|
| 3 |
+
"apply_layernorm": true,
|
| 4 |
+
"architectures": [
|
| 5 |
+
"Dinov2ForImageClassification"
|
| 6 |
+
],
|
| 7 |
+
"attention_probs_dropout_prob": 0.0,
|
| 8 |
+
"drop_path_rate": 0.0,
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.0,
|
| 11 |
+
"hidden_size": 1024,
|
| 12 |
+
"id2label": {
|
| 13 |
+
"0": "correct",
|
| 14 |
+
"1": "incorrect"
|
| 15 |
+
},
|
| 16 |
+
"image_size": 518,
|
| 17 |
+
"initializer_range": 0.02,
|
| 18 |
+
"label2id": {
|
| 19 |
+
"correct": 0,
|
| 20 |
+
"incorrect": 1
|
| 21 |
+
},
|
| 22 |
+
"layer_norm_eps": 1e-06,
|
| 23 |
+
"layerscale_value": 1.0,
|
| 24 |
+
"mlp_ratio": 4,
|
| 25 |
+
"model_type": "dinov2",
|
| 26 |
+
"num_attention_heads": 16,
|
| 27 |
+
"num_channels": 3,
|
| 28 |
+
"num_hidden_layers": 24,
|
| 29 |
+
"out_features": [
|
| 30 |
+
"stage24"
|
| 31 |
+
],
|
| 32 |
+
"out_indices": [
|
| 33 |
+
24
|
| 34 |
+
],
|
| 35 |
+
"patch_size": 14,
|
| 36 |
+
"problem_type": "single_label_classification",
|
| 37 |
+
"qkv_bias": true,
|
| 38 |
+
"reshape_hidden_states": true,
|
| 39 |
+
"stage_names": [
|
| 40 |
+
"stem",
|
| 41 |
+
"stage1",
|
| 42 |
+
"stage2",
|
| 43 |
+
"stage3",
|
| 44 |
+
"stage4",
|
| 45 |
+
"stage5",
|
| 46 |
+
"stage6",
|
| 47 |
+
"stage7",
|
| 48 |
+
"stage8",
|
| 49 |
+
"stage9",
|
| 50 |
+
"stage10",
|
| 51 |
+
"stage11",
|
| 52 |
+
"stage12",
|
| 53 |
+
"stage13",
|
| 54 |
+
"stage14",
|
| 55 |
+
"stage15",
|
| 56 |
+
"stage16",
|
| 57 |
+
"stage17",
|
| 58 |
+
"stage18",
|
| 59 |
+
"stage19",
|
| 60 |
+
"stage20",
|
| 61 |
+
"stage21",
|
| 62 |
+
"stage22",
|
| 63 |
+
"stage23",
|
| 64 |
+
"stage24"
|
| 65 |
+
],
|
| 66 |
+
"torch_dtype": "float32",
|
| 67 |
+
"transformers_version": "4.47.1",
|
| 68 |
+
"use_swiglu_ffn": false
|
| 69 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c4db45e1375a111330f445f36615ec59a496219282a40ba73885298d28e14472
|
| 3 |
+
size 1217542512
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4d60761ac8c8b5a161ea9b8f512bb1924d67f201c99d9adca3a338bf88aac23f
|
| 3 |
+
size 2435341946
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"crop_size": {
|
| 3 |
+
"height": 224,
|
| 4 |
+
"width": 224
|
| 5 |
+
},
|
| 6 |
+
"do_center_crop": true,
|
| 7 |
+
"do_convert_rgb": true,
|
| 8 |
+
"do_normalize": true,
|
| 9 |
+
"do_rescale": true,
|
| 10 |
+
"do_resize": true,
|
| 11 |
+
"image_mean": [
|
| 12 |
+
0.485,
|
| 13 |
+
0.456,
|
| 14 |
+
0.406
|
| 15 |
+
],
|
| 16 |
+
"image_processor_type": "BitImageProcessor",
|
| 17 |
+
"image_std": [
|
| 18 |
+
0.229,
|
| 19 |
+
0.224,
|
| 20 |
+
0.225
|
| 21 |
+
],
|
| 22 |
+
"resample": 3,
|
| 23 |
+
"rescale_factor": 0.00392156862745098,
|
| 24 |
+
"size": {
|
| 25 |
+
"shortest_edge": 256
|
| 26 |
+
}
|
| 27 |
+
}
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:660ccec1688742a08ac50147863109a12b04472562e401ba83f158155084b971
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7fc364201fbceac17ee28f0c25f6cd2003904f59d3a8ddb2469ba7bfdd346578
|
| 3 |
+
size 1064
|
trainer_state.json
ADDED
|
@@ -0,0 +1,875 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9417692129092176,
|
| 3 |
+
"best_model_checkpoint": "Crosswalk/dinov2/checkpoint-924",
|
| 4 |
+
"epoch": 22.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 924,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.24242424242424243,
|
| 13 |
+
"grad_norm": 1809.6781005859375,
|
| 14 |
+
"learning_rate": 9.70873786407767e-07,
|
| 15 |
+
"loss": 4.7087,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.48484848484848486,
|
| 20 |
+
"grad_norm": 190.5601806640625,
|
| 21 |
+
"learning_rate": 1.941747572815534e-06,
|
| 22 |
+
"loss": 3.034,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.7272727272727273,
|
| 27 |
+
"grad_norm": 76.29146575927734,
|
| 28 |
+
"learning_rate": 2.912621359223301e-06,
|
| 29 |
+
"loss": 2.0024,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.9696969696969697,
|
| 34 |
+
"grad_norm": 75.1009750366211,
|
| 35 |
+
"learning_rate": 3.883495145631068e-06,
|
| 36 |
+
"loss": 1.4019,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 1.0,
|
| 41 |
+
"eval_loss": 0.2978227138519287,
|
| 42 |
+
"eval_macro_f1": 0.8815037150933147,
|
| 43 |
+
"eval_runtime": 7.5247,
|
| 44 |
+
"eval_samples_per_second": 43.856,
|
| 45 |
+
"eval_steps_per_second": 5.582,
|
| 46 |
+
"step": 42
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 1.1939393939393939,
|
| 50 |
+
"grad_norm": 86.90376281738281,
|
| 51 |
+
"learning_rate": 4.854368932038836e-06,
|
| 52 |
+
"loss": 0.7179,
|
| 53 |
+
"step": 50
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.4363636363636363,
|
| 57 |
+
"grad_norm": 118.144287109375,
|
| 58 |
+
"learning_rate": 5.825242718446602e-06,
|
| 59 |
+
"loss": 0.9737,
|
| 60 |
+
"step": 60
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 1.6787878787878787,
|
| 64 |
+
"grad_norm": 174.987548828125,
|
| 65 |
+
"learning_rate": 6.79611650485437e-06,
|
| 66 |
+
"loss": 1.3976,
|
| 67 |
+
"step": 70
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 1.9212121212121214,
|
| 71 |
+
"grad_norm": 35.86643981933594,
|
| 72 |
+
"learning_rate": 7.766990291262136e-06,
|
| 73 |
+
"loss": 0.967,
|
| 74 |
+
"step": 80
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 2.0,
|
| 78 |
+
"eval_loss": 0.22084859013557434,
|
| 79 |
+
"eval_macro_f1": 0.9229339286881953,
|
| 80 |
+
"eval_runtime": 5.8751,
|
| 81 |
+
"eval_samples_per_second": 56.169,
|
| 82 |
+
"eval_steps_per_second": 7.149,
|
| 83 |
+
"step": 84
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 2.1454545454545455,
|
| 87 |
+
"grad_norm": 232.57371520996094,
|
| 88 |
+
"learning_rate": 8.737864077669904e-06,
|
| 89 |
+
"loss": 1.323,
|
| 90 |
+
"step": 90
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 2.3878787878787877,
|
| 94 |
+
"grad_norm": 72.19599914550781,
|
| 95 |
+
"learning_rate": 9.708737864077671e-06,
|
| 96 |
+
"loss": 1.005,
|
| 97 |
+
"step": 100
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 2.6303030303030304,
|
| 101 |
+
"grad_norm": 55.63515853881836,
|
| 102 |
+
"learning_rate": 9.924078091106291e-06,
|
| 103 |
+
"loss": 0.6989,
|
| 104 |
+
"step": 110
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 2.8727272727272726,
|
| 108 |
+
"grad_norm": 56.631591796875,
|
| 109 |
+
"learning_rate": 9.815618221258135e-06,
|
| 110 |
+
"loss": 0.7527,
|
| 111 |
+
"step": 120
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 3.0,
|
| 115 |
+
"eval_loss": 0.3025018870830536,
|
| 116 |
+
"eval_macro_f1": 0.9009343690194753,
|
| 117 |
+
"eval_runtime": 6.0811,
|
| 118 |
+
"eval_samples_per_second": 54.266,
|
| 119 |
+
"eval_steps_per_second": 6.907,
|
| 120 |
+
"step": 126
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 3.096969696969697,
|
| 124 |
+
"grad_norm": 40.3350715637207,
|
| 125 |
+
"learning_rate": 9.70715835140998e-06,
|
| 126 |
+
"loss": 0.6221,
|
| 127 |
+
"step": 130
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 3.3393939393939394,
|
| 131 |
+
"grad_norm": 95.85454559326172,
|
| 132 |
+
"learning_rate": 9.598698481561823e-06,
|
| 133 |
+
"loss": 0.9071,
|
| 134 |
+
"step": 140
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 3.581818181818182,
|
| 138 |
+
"grad_norm": 94.18701934814453,
|
| 139 |
+
"learning_rate": 9.490238611713667e-06,
|
| 140 |
+
"loss": 0.7042,
|
| 141 |
+
"step": 150
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 3.824242424242424,
|
| 145 |
+
"grad_norm": 477.8069763183594,
|
| 146 |
+
"learning_rate": 9.38177874186551e-06,
|
| 147 |
+
"loss": 0.635,
|
| 148 |
+
"step": 160
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 4.0,
|
| 152 |
+
"eval_loss": 0.22365985810756683,
|
| 153 |
+
"eval_macro_f1": 0.9049918736939866,
|
| 154 |
+
"eval_runtime": 5.9294,
|
| 155 |
+
"eval_samples_per_second": 55.655,
|
| 156 |
+
"eval_steps_per_second": 7.083,
|
| 157 |
+
"step": 168
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 4.048484848484849,
|
| 161 |
+
"grad_norm": 110.27919006347656,
|
| 162 |
+
"learning_rate": 9.273318872017354e-06,
|
| 163 |
+
"loss": 1.1381,
|
| 164 |
+
"step": 170
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 4.290909090909091,
|
| 168 |
+
"grad_norm": 58.735958099365234,
|
| 169 |
+
"learning_rate": 9.1648590021692e-06,
|
| 170 |
+
"loss": 0.7225,
|
| 171 |
+
"step": 180
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 4.533333333333333,
|
| 175 |
+
"grad_norm": 75.20926666259766,
|
| 176 |
+
"learning_rate": 9.056399132321042e-06,
|
| 177 |
+
"loss": 0.4634,
|
| 178 |
+
"step": 190
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 4.775757575757575,
|
| 182 |
+
"grad_norm": 18.876136779785156,
|
| 183 |
+
"learning_rate": 8.947939262472886e-06,
|
| 184 |
+
"loss": 0.6293,
|
| 185 |
+
"step": 200
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 5.0,
|
| 189 |
+
"grad_norm": 8.281109809875488,
|
| 190 |
+
"learning_rate": 8.83947939262473e-06,
|
| 191 |
+
"loss": 0.6632,
|
| 192 |
+
"step": 210
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 5.0,
|
| 196 |
+
"eval_loss": 0.2299780696630478,
|
| 197 |
+
"eval_macro_f1": 0.9176304185040354,
|
| 198 |
+
"eval_runtime": 6.0281,
|
| 199 |
+
"eval_samples_per_second": 54.744,
|
| 200 |
+
"eval_steps_per_second": 6.967,
|
| 201 |
+
"step": 210
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 5.242424242424242,
|
| 205 |
+
"grad_norm": 13.756321907043457,
|
| 206 |
+
"learning_rate": 8.731019522776574e-06,
|
| 207 |
+
"loss": 0.4708,
|
| 208 |
+
"step": 220
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 5.484848484848484,
|
| 212 |
+
"grad_norm": 59.22605895996094,
|
| 213 |
+
"learning_rate": 8.622559652928418e-06,
|
| 214 |
+
"loss": 0.7127,
|
| 215 |
+
"step": 230
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 5.7272727272727275,
|
| 219 |
+
"grad_norm": 32.43043899536133,
|
| 220 |
+
"learning_rate": 8.514099783080262e-06,
|
| 221 |
+
"loss": 0.5682,
|
| 222 |
+
"step": 240
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 5.96969696969697,
|
| 226 |
+
"grad_norm": 54.722599029541016,
|
| 227 |
+
"learning_rate": 8.405639913232104e-06,
|
| 228 |
+
"loss": 0.8667,
|
| 229 |
+
"step": 250
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 6.0,
|
| 233 |
+
"eval_loss": 0.2767850160598755,
|
| 234 |
+
"eval_macro_f1": 0.9210700618192522,
|
| 235 |
+
"eval_runtime": 6.2004,
|
| 236 |
+
"eval_samples_per_second": 53.223,
|
| 237 |
+
"eval_steps_per_second": 6.774,
|
| 238 |
+
"step": 252
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 6.193939393939394,
|
| 242 |
+
"grad_norm": 18.175823211669922,
|
| 243 |
+
"learning_rate": 8.29718004338395e-06,
|
| 244 |
+
"loss": 0.6752,
|
| 245 |
+
"step": 260
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 6.4363636363636365,
|
| 249 |
+
"grad_norm": 56.679988861083984,
|
| 250 |
+
"learning_rate": 8.188720173535792e-06,
|
| 251 |
+
"loss": 0.3727,
|
| 252 |
+
"step": 270
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 6.678787878787879,
|
| 256 |
+
"grad_norm": 57.3917236328125,
|
| 257 |
+
"learning_rate": 8.080260303687636e-06,
|
| 258 |
+
"loss": 1.0167,
|
| 259 |
+
"step": 280
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 6.921212121212121,
|
| 263 |
+
"grad_norm": 42.186038970947266,
|
| 264 |
+
"learning_rate": 7.97180043383948e-06,
|
| 265 |
+
"loss": 0.9377,
|
| 266 |
+
"step": 290
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 7.0,
|
| 270 |
+
"eval_loss": 0.29274508357048035,
|
| 271 |
+
"eval_macro_f1": 0.9138863000931967,
|
| 272 |
+
"eval_runtime": 6.1213,
|
| 273 |
+
"eval_samples_per_second": 53.91,
|
| 274 |
+
"eval_steps_per_second": 6.861,
|
| 275 |
+
"step": 294
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 7.1454545454545455,
|
| 279 |
+
"grad_norm": 41.31782531738281,
|
| 280 |
+
"learning_rate": 7.863340563991324e-06,
|
| 281 |
+
"loss": 0.3818,
|
| 282 |
+
"step": 300
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 7.387878787878788,
|
| 286 |
+
"grad_norm": 4.223178863525391,
|
| 287 |
+
"learning_rate": 7.754880694143168e-06,
|
| 288 |
+
"loss": 0.4503,
|
| 289 |
+
"step": 310
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 7.63030303030303,
|
| 293 |
+
"grad_norm": 35.64258575439453,
|
| 294 |
+
"learning_rate": 7.646420824295012e-06,
|
| 295 |
+
"loss": 0.6038,
|
| 296 |
+
"step": 320
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 7.872727272727273,
|
| 300 |
+
"grad_norm": 37.91206359863281,
|
| 301 |
+
"learning_rate": 7.537960954446856e-06,
|
| 302 |
+
"loss": 0.5407,
|
| 303 |
+
"step": 330
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 8.0,
|
| 307 |
+
"eval_loss": 0.20143219828605652,
|
| 308 |
+
"eval_macro_f1": 0.9357970705676355,
|
| 309 |
+
"eval_runtime": 5.9715,
|
| 310 |
+
"eval_samples_per_second": 55.263,
|
| 311 |
+
"eval_steps_per_second": 7.033,
|
| 312 |
+
"step": 336
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 8.096969696969698,
|
| 316 |
+
"grad_norm": 9.571391105651855,
|
| 317 |
+
"learning_rate": 7.429501084598699e-06,
|
| 318 |
+
"loss": 0.3311,
|
| 319 |
+
"step": 340
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 8.33939393939394,
|
| 323 |
+
"grad_norm": 30.14655876159668,
|
| 324 |
+
"learning_rate": 7.321041214750543e-06,
|
| 325 |
+
"loss": 0.5367,
|
| 326 |
+
"step": 350
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 8.581818181818182,
|
| 330 |
+
"grad_norm": 125.38350677490234,
|
| 331 |
+
"learning_rate": 7.212581344902386e-06,
|
| 332 |
+
"loss": 0.4511,
|
| 333 |
+
"step": 360
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 8.824242424242424,
|
| 337 |
+
"grad_norm": 283.20819091796875,
|
| 338 |
+
"learning_rate": 7.104121475054231e-06,
|
| 339 |
+
"loss": 0.5474,
|
| 340 |
+
"step": 370
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 9.0,
|
| 344 |
+
"eval_loss": 0.329227477312088,
|
| 345 |
+
"eval_macro_f1": 0.8817302125547928,
|
| 346 |
+
"eval_runtime": 5.984,
|
| 347 |
+
"eval_samples_per_second": 55.147,
|
| 348 |
+
"eval_steps_per_second": 7.019,
|
| 349 |
+
"step": 378
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 9.048484848484849,
|
| 353 |
+
"grad_norm": 31.927379608154297,
|
| 354 |
+
"learning_rate": 6.995661605206075e-06,
|
| 355 |
+
"loss": 0.3963,
|
| 356 |
+
"step": 380
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 9.290909090909091,
|
| 360 |
+
"grad_norm": 10.10098934173584,
|
| 361 |
+
"learning_rate": 6.887201735357918e-06,
|
| 362 |
+
"loss": 0.445,
|
| 363 |
+
"step": 390
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 9.533333333333333,
|
| 367 |
+
"grad_norm": 1.947770118713379,
|
| 368 |
+
"learning_rate": 6.778741865509761e-06,
|
| 369 |
+
"loss": 0.5947,
|
| 370 |
+
"step": 400
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 9.775757575757575,
|
| 374 |
+
"grad_norm": 54.11802673339844,
|
| 375 |
+
"learning_rate": 6.670281995661606e-06,
|
| 376 |
+
"loss": 0.6001,
|
| 377 |
+
"step": 410
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 10.0,
|
| 381 |
+
"grad_norm": 0.004518165718764067,
|
| 382 |
+
"learning_rate": 6.56182212581345e-06,
|
| 383 |
+
"loss": 0.412,
|
| 384 |
+
"step": 420
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 10.0,
|
| 388 |
+
"eval_loss": 0.3594599962234497,
|
| 389 |
+
"eval_macro_f1": 0.907735321528425,
|
| 390 |
+
"eval_runtime": 6.0125,
|
| 391 |
+
"eval_samples_per_second": 54.885,
|
| 392 |
+
"eval_steps_per_second": 6.985,
|
| 393 |
+
"step": 420
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 10.242424242424242,
|
| 397 |
+
"grad_norm": 0.22337216138839722,
|
| 398 |
+
"learning_rate": 6.453362255965293e-06,
|
| 399 |
+
"loss": 0.2798,
|
| 400 |
+
"step": 430
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 10.484848484848484,
|
| 404 |
+
"grad_norm": 39.639984130859375,
|
| 405 |
+
"learning_rate": 6.344902386117138e-06,
|
| 406 |
+
"loss": 0.4377,
|
| 407 |
+
"step": 440
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 10.727272727272727,
|
| 411 |
+
"grad_norm": 53.85198211669922,
|
| 412 |
+
"learning_rate": 6.236442516268981e-06,
|
| 413 |
+
"loss": 0.2063,
|
| 414 |
+
"step": 450
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 10.969696969696969,
|
| 418 |
+
"grad_norm": 69.08942413330078,
|
| 419 |
+
"learning_rate": 6.127982646420825e-06,
|
| 420 |
+
"loss": 0.2884,
|
| 421 |
+
"step": 460
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 11.0,
|
| 425 |
+
"eval_loss": 0.2930862307548523,
|
| 426 |
+
"eval_macro_f1": 0.9380839806371721,
|
| 427 |
+
"eval_runtime": 6.0147,
|
| 428 |
+
"eval_samples_per_second": 54.866,
|
| 429 |
+
"eval_steps_per_second": 6.983,
|
| 430 |
+
"step": 462
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 11.193939393939393,
|
| 434 |
+
"grad_norm": 63.143218994140625,
|
| 435 |
+
"learning_rate": 6.019522776572668e-06,
|
| 436 |
+
"loss": 0.6075,
|
| 437 |
+
"step": 470
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 11.436363636363636,
|
| 441 |
+
"grad_norm": 7.950187683105469,
|
| 442 |
+
"learning_rate": 5.911062906724513e-06,
|
| 443 |
+
"loss": 0.2654,
|
| 444 |
+
"step": 480
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 11.67878787878788,
|
| 448 |
+
"grad_norm": 82.30758666992188,
|
| 449 |
+
"learning_rate": 5.802603036876356e-06,
|
| 450 |
+
"loss": 0.2474,
|
| 451 |
+
"step": 490
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 11.921212121212122,
|
| 455 |
+
"grad_norm": 7.340689182281494,
|
| 456 |
+
"learning_rate": 5.6941431670282e-06,
|
| 457 |
+
"loss": 0.2405,
|
| 458 |
+
"step": 500
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 12.0,
|
| 462 |
+
"eval_loss": 0.3316686451435089,
|
| 463 |
+
"eval_macro_f1": 0.9209216589861751,
|
| 464 |
+
"eval_runtime": 5.8916,
|
| 465 |
+
"eval_samples_per_second": 56.012,
|
| 466 |
+
"eval_steps_per_second": 7.129,
|
| 467 |
+
"step": 504
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 12.145454545454545,
|
| 471 |
+
"grad_norm": 37.141632080078125,
|
| 472 |
+
"learning_rate": 5.585683297180043e-06,
|
| 473 |
+
"loss": 0.3349,
|
| 474 |
+
"step": 510
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 12.387878787878789,
|
| 478 |
+
"grad_norm": 34.024383544921875,
|
| 479 |
+
"learning_rate": 5.477223427331888e-06,
|
| 480 |
+
"loss": 0.1742,
|
| 481 |
+
"step": 520
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 12.63030303030303,
|
| 485 |
+
"grad_norm": 46.10781478881836,
|
| 486 |
+
"learning_rate": 5.368763557483731e-06,
|
| 487 |
+
"loss": 0.2115,
|
| 488 |
+
"step": 530
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 12.872727272727273,
|
| 492 |
+
"grad_norm": 126.67015838623047,
|
| 493 |
+
"learning_rate": 5.260303687635575e-06,
|
| 494 |
+
"loss": 0.8788,
|
| 495 |
+
"step": 540
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 13.0,
|
| 499 |
+
"eval_loss": 0.37741926312446594,
|
| 500 |
+
"eval_macro_f1": 0.9058106453305834,
|
| 501 |
+
"eval_runtime": 6.6329,
|
| 502 |
+
"eval_samples_per_second": 49.752,
|
| 503 |
+
"eval_steps_per_second": 6.332,
|
| 504 |
+
"step": 546
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 13.096969696969698,
|
| 508 |
+
"grad_norm": 35.44662094116211,
|
| 509 |
+
"learning_rate": 5.151843817787418e-06,
|
| 510 |
+
"loss": 0.5591,
|
| 511 |
+
"step": 550
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 13.33939393939394,
|
| 515 |
+
"grad_norm": 57.34544372558594,
|
| 516 |
+
"learning_rate": 5.043383947939263e-06,
|
| 517 |
+
"loss": 0.213,
|
| 518 |
+
"step": 560
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 13.581818181818182,
|
| 522 |
+
"grad_norm": 15.285223960876465,
|
| 523 |
+
"learning_rate": 4.934924078091107e-06,
|
| 524 |
+
"loss": 0.203,
|
| 525 |
+
"step": 570
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 13.824242424242424,
|
| 529 |
+
"grad_norm": 28.99003028869629,
|
| 530 |
+
"learning_rate": 4.82646420824295e-06,
|
| 531 |
+
"loss": 0.4163,
|
| 532 |
+
"step": 580
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 14.0,
|
| 536 |
+
"eval_loss": 0.39865490794181824,
|
| 537 |
+
"eval_macro_f1": 0.9196508840275697,
|
| 538 |
+
"eval_runtime": 5.8701,
|
| 539 |
+
"eval_samples_per_second": 56.217,
|
| 540 |
+
"eval_steps_per_second": 7.155,
|
| 541 |
+
"step": 588
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 14.048484848484849,
|
| 545 |
+
"grad_norm": 346.8255310058594,
|
| 546 |
+
"learning_rate": 4.718004338394794e-06,
|
| 547 |
+
"loss": 0.26,
|
| 548 |
+
"step": 590
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 14.290909090909091,
|
| 552 |
+
"grad_norm": 38.04654312133789,
|
| 553 |
+
"learning_rate": 4.609544468546638e-06,
|
| 554 |
+
"loss": 0.3813,
|
| 555 |
+
"step": 600
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 14.533333333333333,
|
| 559 |
+
"grad_norm": 34.71643829345703,
|
| 560 |
+
"learning_rate": 4.501084598698482e-06,
|
| 561 |
+
"loss": 0.0974,
|
| 562 |
+
"step": 610
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 14.775757575757575,
|
| 566 |
+
"grad_norm": 34.031890869140625,
|
| 567 |
+
"learning_rate": 4.392624728850326e-06,
|
| 568 |
+
"loss": 0.4881,
|
| 569 |
+
"step": 620
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 15.0,
|
| 573 |
+
"grad_norm": 0.0002771662548184395,
|
| 574 |
+
"learning_rate": 4.284164859002169e-06,
|
| 575 |
+
"loss": 0.4126,
|
| 576 |
+
"step": 630
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 15.0,
|
| 580 |
+
"eval_loss": 0.35451531410217285,
|
| 581 |
+
"eval_macro_f1": 0.9235679411519468,
|
| 582 |
+
"eval_runtime": 6.0428,
|
| 583 |
+
"eval_samples_per_second": 54.611,
|
| 584 |
+
"eval_steps_per_second": 6.95,
|
| 585 |
+
"step": 630
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 15.242424242424242,
|
| 589 |
+
"grad_norm": 82.48748779296875,
|
| 590 |
+
"learning_rate": 4.175704989154013e-06,
|
| 591 |
+
"loss": 0.4444,
|
| 592 |
+
"step": 640
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 15.484848484848484,
|
| 596 |
+
"grad_norm": 0.2618753910064697,
|
| 597 |
+
"learning_rate": 4.067245119305857e-06,
|
| 598 |
+
"loss": 0.2083,
|
| 599 |
+
"step": 650
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 15.727272727272727,
|
| 603 |
+
"grad_norm": 98.34405517578125,
|
| 604 |
+
"learning_rate": 3.958785249457701e-06,
|
| 605 |
+
"loss": 0.4785,
|
| 606 |
+
"step": 660
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 15.969696969696969,
|
| 610 |
+
"grad_norm": 0.37142229080200195,
|
| 611 |
+
"learning_rate": 3.8503253796095445e-06,
|
| 612 |
+
"loss": 0.1583,
|
| 613 |
+
"step": 670
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 16.0,
|
| 617 |
+
"eval_loss": 0.38117873668670654,
|
| 618 |
+
"eval_macro_f1": 0.9268860086407444,
|
| 619 |
+
"eval_runtime": 6.9311,
|
| 620 |
+
"eval_samples_per_second": 47.612,
|
| 621 |
+
"eval_steps_per_second": 6.06,
|
| 622 |
+
"step": 672
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 16.193939393939395,
|
| 626 |
+
"grad_norm": 40.54330825805664,
|
| 627 |
+
"learning_rate": 3.741865509761389e-06,
|
| 628 |
+
"loss": 0.0774,
|
| 629 |
+
"step": 680
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 16.436363636363637,
|
| 633 |
+
"grad_norm": 0.23675695061683655,
|
| 634 |
+
"learning_rate": 3.6334056399132324e-06,
|
| 635 |
+
"loss": 0.1639,
|
| 636 |
+
"step": 690
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 16.67878787878788,
|
| 640 |
+
"grad_norm": 47.12529373168945,
|
| 641 |
+
"learning_rate": 3.5249457700650764e-06,
|
| 642 |
+
"loss": 0.306,
|
| 643 |
+
"step": 700
|
| 644 |
+
},
|
| 645 |
+
{
|
| 646 |
+
"epoch": 16.921212121212122,
|
| 647 |
+
"grad_norm": 0.3993530571460724,
|
| 648 |
+
"learning_rate": 3.41648590021692e-06,
|
| 649 |
+
"loss": 0.2376,
|
| 650 |
+
"step": 710
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 17.0,
|
| 654 |
+
"eval_loss": 0.4087267816066742,
|
| 655 |
+
"eval_macro_f1": 0.9295990205081115,
|
| 656 |
+
"eval_runtime": 6.1306,
|
| 657 |
+
"eval_samples_per_second": 53.828,
|
| 658 |
+
"eval_steps_per_second": 6.851,
|
| 659 |
+
"step": 714
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 17.145454545454545,
|
| 663 |
+
"grad_norm": 0.34205177426338196,
|
| 664 |
+
"learning_rate": 3.308026030368764e-06,
|
| 665 |
+
"loss": 0.0332,
|
| 666 |
+
"step": 720
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 17.387878787878787,
|
| 670 |
+
"grad_norm": 0.5112647414207458,
|
| 671 |
+
"learning_rate": 3.1995661605206075e-06,
|
| 672 |
+
"loss": 0.1332,
|
| 673 |
+
"step": 730
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"epoch": 17.63030303030303,
|
| 677 |
+
"grad_norm": 120.2950439453125,
|
| 678 |
+
"learning_rate": 3.0911062906724515e-06,
|
| 679 |
+
"loss": 0.2503,
|
| 680 |
+
"step": 740
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 17.87272727272727,
|
| 684 |
+
"grad_norm": 223.04759216308594,
|
| 685 |
+
"learning_rate": 2.982646420824295e-06,
|
| 686 |
+
"loss": 0.2703,
|
| 687 |
+
"step": 750
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 18.0,
|
| 691 |
+
"eval_loss": 0.43362897634506226,
|
| 692 |
+
"eval_macro_f1": 0.9264924264924266,
|
| 693 |
+
"eval_runtime": 5.861,
|
| 694 |
+
"eval_samples_per_second": 56.305,
|
| 695 |
+
"eval_steps_per_second": 7.166,
|
| 696 |
+
"step": 756
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 18.096969696969698,
|
| 700 |
+
"grad_norm": 54.66193771362305,
|
| 701 |
+
"learning_rate": 2.874186550976139e-06,
|
| 702 |
+
"loss": 0.1274,
|
| 703 |
+
"step": 760
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 18.33939393939394,
|
| 707 |
+
"grad_norm": 54.846466064453125,
|
| 708 |
+
"learning_rate": 2.765726681127983e-06,
|
| 709 |
+
"loss": 0.2751,
|
| 710 |
+
"step": 770
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 18.581818181818182,
|
| 714 |
+
"grad_norm": 53.97863006591797,
|
| 715 |
+
"learning_rate": 2.6572668112798266e-06,
|
| 716 |
+
"loss": 0.359,
|
| 717 |
+
"step": 780
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 18.824242424242424,
|
| 721 |
+
"grad_norm": 81.63549041748047,
|
| 722 |
+
"learning_rate": 2.5488069414316706e-06,
|
| 723 |
+
"loss": 0.1819,
|
| 724 |
+
"step": 790
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 19.0,
|
| 728 |
+
"eval_loss": 0.3480012118816376,
|
| 729 |
+
"eval_macro_f1": 0.9236528192931639,
|
| 730 |
+
"eval_runtime": 7.0471,
|
| 731 |
+
"eval_samples_per_second": 46.828,
|
| 732 |
+
"eval_steps_per_second": 5.96,
|
| 733 |
+
"step": 798
|
| 734 |
+
},
|
| 735 |
+
{
|
| 736 |
+
"epoch": 19.048484848484847,
|
| 737 |
+
"grad_norm": 41.54087448120117,
|
| 738 |
+
"learning_rate": 2.440347071583514e-06,
|
| 739 |
+
"loss": 0.6373,
|
| 740 |
+
"step": 800
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 19.29090909090909,
|
| 744 |
+
"grad_norm": 9.001028060913086,
|
| 745 |
+
"learning_rate": 2.331887201735358e-06,
|
| 746 |
+
"loss": 0.2971,
|
| 747 |
+
"step": 810
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 19.533333333333335,
|
| 751 |
+
"grad_norm": 114.6279525756836,
|
| 752 |
+
"learning_rate": 2.2234273318872017e-06,
|
| 753 |
+
"loss": 0.1943,
|
| 754 |
+
"step": 820
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 19.775757575757577,
|
| 758 |
+
"grad_norm": 18.022676467895508,
|
| 759 |
+
"learning_rate": 2.1149674620390457e-06,
|
| 760 |
+
"loss": 0.1207,
|
| 761 |
+
"step": 830
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 20.0,
|
| 765 |
+
"grad_norm": 0.0001882202341221273,
|
| 766 |
+
"learning_rate": 2.0065075921908892e-06,
|
| 767 |
+
"loss": 0.1324,
|
| 768 |
+
"step": 840
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 20.0,
|
| 772 |
+
"eval_loss": 0.4493299424648285,
|
| 773 |
+
"eval_macro_f1": 0.9384902143522833,
|
| 774 |
+
"eval_runtime": 6.0147,
|
| 775 |
+
"eval_samples_per_second": 54.865,
|
| 776 |
+
"eval_steps_per_second": 6.983,
|
| 777 |
+
"step": 840
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 20.242424242424242,
|
| 781 |
+
"grad_norm": 13.740226745605469,
|
| 782 |
+
"learning_rate": 1.8980477223427332e-06,
|
| 783 |
+
"loss": 0.294,
|
| 784 |
+
"step": 850
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 20.484848484848484,
|
| 788 |
+
"grad_norm": 6.870513916015625,
|
| 789 |
+
"learning_rate": 1.7895878524945772e-06,
|
| 790 |
+
"loss": 0.1323,
|
| 791 |
+
"step": 860
|
| 792 |
+
},
|
| 793 |
+
{
|
| 794 |
+
"epoch": 20.727272727272727,
|
| 795 |
+
"grad_norm": 2.74729585647583,
|
| 796 |
+
"learning_rate": 1.681127982646421e-06,
|
| 797 |
+
"loss": 0.019,
|
| 798 |
+
"step": 870
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"epoch": 20.96969696969697,
|
| 802 |
+
"grad_norm": 17.18338966369629,
|
| 803 |
+
"learning_rate": 1.572668112798265e-06,
|
| 804 |
+
"loss": 0.1312,
|
| 805 |
+
"step": 880
|
| 806 |
+
},
|
| 807 |
+
{
|
| 808 |
+
"epoch": 21.0,
|
| 809 |
+
"eval_loss": 0.40448498725891113,
|
| 810 |
+
"eval_macro_f1": 0.9384902143522833,
|
| 811 |
+
"eval_runtime": 6.133,
|
| 812 |
+
"eval_samples_per_second": 53.808,
|
| 813 |
+
"eval_steps_per_second": 6.848,
|
| 814 |
+
"step": 882
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 21.193939393939395,
|
| 818 |
+
"grad_norm": 108.9833755493164,
|
| 819 |
+
"learning_rate": 1.4642082429501087e-06,
|
| 820 |
+
"loss": 0.2499,
|
| 821 |
+
"step": 890
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 21.436363636363637,
|
| 825 |
+
"grad_norm": 0.291847825050354,
|
| 826 |
+
"learning_rate": 1.3557483731019525e-06,
|
| 827 |
+
"loss": 0.1708,
|
| 828 |
+
"step": 900
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 21.67878787878788,
|
| 832 |
+
"grad_norm": 35.08168029785156,
|
| 833 |
+
"learning_rate": 1.2472885032537963e-06,
|
| 834 |
+
"loss": 0.0802,
|
| 835 |
+
"step": 910
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 21.921212121212122,
|
| 839 |
+
"grad_norm": 0.05401836335659027,
|
| 840 |
+
"learning_rate": 1.13882863340564e-06,
|
| 841 |
+
"loss": 0.1662,
|
| 842 |
+
"step": 920
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 22.0,
|
| 846 |
+
"eval_loss": 0.3166828453540802,
|
| 847 |
+
"eval_macro_f1": 0.9417692129092176,
|
| 848 |
+
"eval_runtime": 6.0442,
|
| 849 |
+
"eval_samples_per_second": 54.598,
|
| 850 |
+
"eval_steps_per_second": 6.949,
|
| 851 |
+
"step": 924
|
| 852 |
+
}
|
| 853 |
+
],
|
| 854 |
+
"logging_steps": 10,
|
| 855 |
+
"max_steps": 1025,
|
| 856 |
+
"num_input_tokens_seen": 0,
|
| 857 |
+
"num_train_epochs": 25,
|
| 858 |
+
"save_steps": 500,
|
| 859 |
+
"stateful_callbacks": {
|
| 860 |
+
"TrainerControl": {
|
| 861 |
+
"args": {
|
| 862 |
+
"should_epoch_stop": false,
|
| 863 |
+
"should_evaluate": false,
|
| 864 |
+
"should_log": false,
|
| 865 |
+
"should_save": true,
|
| 866 |
+
"should_training_stop": false
|
| 867 |
+
},
|
| 868 |
+
"attributes": {}
|
| 869 |
+
}
|
| 870 |
+
},
|
| 871 |
+
"total_flos": 1.0371596050603966e+19,
|
| 872 |
+
"train_batch_size": 8,
|
| 873 |
+
"trial_name": null,
|
| 874 |
+
"trial_params": null
|
| 875 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d0b1d45c6c028d437456c5af083b3618770c1bf92ea996d351710771438e3073
|
| 3 |
+
size 5304
|