Training in progress, step 15000, checkpoint
Browse files- checkpoint-15000/config.json +62 -0
- checkpoint-15000/model.safetensors +3 -0
- checkpoint-15000/optimizer.pt +3 -0
- checkpoint-15000/preprocessor_config.json +26 -0
- checkpoint-15000/rng_state.pth +3 -0
- checkpoint-15000/scheduler.pt +3 -0
- checkpoint-15000/trainer_state.json +2283 -0
- checkpoint-15000/training_args.bin +3 -0
checkpoint-15000/config.json
ADDED
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "facebook/detr-resnet-50-dc5",
|
3 |
+
"activation_dropout": 0.0,
|
4 |
+
"activation_function": "relu",
|
5 |
+
"architectures": [
|
6 |
+
"DetrForObjectDetection"
|
7 |
+
],
|
8 |
+
"attention_dropout": 0.0,
|
9 |
+
"auxiliary_loss": false,
|
10 |
+
"backbone": "resnet50",
|
11 |
+
"backbone_config": null,
|
12 |
+
"backbone_kwargs": {
|
13 |
+
"in_chans": 3,
|
14 |
+
"out_indices": [
|
15 |
+
1,
|
16 |
+
2,
|
17 |
+
3,
|
18 |
+
4
|
19 |
+
],
|
20 |
+
"output_stride": 16
|
21 |
+
},
|
22 |
+
"bbox_cost": 5,
|
23 |
+
"bbox_loss_coefficient": 5,
|
24 |
+
"class_cost": 1,
|
25 |
+
"classifier_dropout": 0.0,
|
26 |
+
"d_model": 256,
|
27 |
+
"decoder_attention_heads": 8,
|
28 |
+
"decoder_ffn_dim": 2048,
|
29 |
+
"decoder_layerdrop": 0.0,
|
30 |
+
"decoder_layers": 6,
|
31 |
+
"dice_loss_coefficient": 1,
|
32 |
+
"dilation": true,
|
33 |
+
"dropout": 0.1,
|
34 |
+
"encoder_attention_heads": 8,
|
35 |
+
"encoder_ffn_dim": 2048,
|
36 |
+
"encoder_layerdrop": 0.0,
|
37 |
+
"encoder_layers": 6,
|
38 |
+
"eos_coefficient": 0.1,
|
39 |
+
"giou_cost": 2,
|
40 |
+
"giou_loss_coefficient": 2,
|
41 |
+
"id2label": {
|
42 |
+
"0": "segment"
|
43 |
+
},
|
44 |
+
"init_std": 0.02,
|
45 |
+
"init_xavier_std": 1.0,
|
46 |
+
"is_encoder_decoder": true,
|
47 |
+
"label2id": {
|
48 |
+
"segment": 0
|
49 |
+
},
|
50 |
+
"mask_loss_coefficient": 1,
|
51 |
+
"max_position_embeddings": 1024,
|
52 |
+
"model_type": "detr",
|
53 |
+
"num_channels": 3,
|
54 |
+
"num_hidden_layers": 6,
|
55 |
+
"num_queries": 500,
|
56 |
+
"position_embedding_type": "sine",
|
57 |
+
"scale_embedding": false,
|
58 |
+
"torch_dtype": "float32",
|
59 |
+
"transformers_version": "4.44.2",
|
60 |
+
"use_pretrained_backbone": true,
|
61 |
+
"use_timm_backbone": true
|
62 |
+
}
|
checkpoint-15000/model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7e3a7550527bfe53766b649b702a7aa467223b10fcba9b32329c8b256d5b0da6
|
3 |
+
size 166904424
|
checkpoint-15000/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6cbd20184ef8ca932bc4f1d3670279a6f3ffbaf724a774eea7e01328bf572b19
|
3 |
+
size 331310970
|
checkpoint-15000/preprocessor_config.json
ADDED
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"do_convert_annotations": true,
|
3 |
+
"do_normalize": true,
|
4 |
+
"do_pad": true,
|
5 |
+
"do_rescale": true,
|
6 |
+
"do_resize": true,
|
7 |
+
"format": "coco_detection",
|
8 |
+
"image_mean": [
|
9 |
+
0.485,
|
10 |
+
0.456,
|
11 |
+
0.406
|
12 |
+
],
|
13 |
+
"image_processor_type": "DetrImageProcessor",
|
14 |
+
"image_std": [
|
15 |
+
0.229,
|
16 |
+
0.224,
|
17 |
+
0.225
|
18 |
+
],
|
19 |
+
"pad_size": null,
|
20 |
+
"resample": 2,
|
21 |
+
"rescale_factor": 0.00392156862745098,
|
22 |
+
"size": {
|
23 |
+
"longest_edge": 1333,
|
24 |
+
"shortest_edge": 800
|
25 |
+
}
|
26 |
+
}
|
checkpoint-15000/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:56be16913f4bf96c1644f694bceb7fc22ece4d749543e93aa97a0a4cc14bb2df
|
3 |
+
size 14244
|
checkpoint-15000/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:330b61f3755f27b04e2954b637e7637df4deaf1ae28e11c5d40d73e7fb2de5c6
|
3 |
+
size 1064
|
checkpoint-15000/trainer_state.json
ADDED
@@ -0,0 +1,2283 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 76.92307692307692,
|
5 |
+
"eval_steps": 100,
|
6 |
+
"global_step": 15000,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.5128205128205128,
|
13 |
+
"grad_norm": 81.8341293334961,
|
14 |
+
"learning_rate": 9.990900000000001e-06,
|
15 |
+
"loss": 3.5606,
|
16 |
+
"step": 100
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.5128205128205128,
|
20 |
+
"eval_loss": 2.2667126655578613,
|
21 |
+
"eval_runtime": 33.7385,
|
22 |
+
"eval_samples_per_second": 11.648,
|
23 |
+
"eval_steps_per_second": 1.482,
|
24 |
+
"step": 100
|
25 |
+
},
|
26 |
+
{
|
27 |
+
"epoch": 1.0256410256410255,
|
28 |
+
"grad_norm": 30.16681671142578,
|
29 |
+
"learning_rate": 9.980900000000001e-06,
|
30 |
+
"loss": 2.1841,
|
31 |
+
"step": 200
|
32 |
+
},
|
33 |
+
{
|
34 |
+
"epoch": 1.0256410256410255,
|
35 |
+
"eval_loss": 2.059685230255127,
|
36 |
+
"eval_runtime": 33.6897,
|
37 |
+
"eval_samples_per_second": 11.665,
|
38 |
+
"eval_steps_per_second": 1.484,
|
39 |
+
"step": 200
|
40 |
+
},
|
41 |
+
{
|
42 |
+
"epoch": 1.5384615384615383,
|
43 |
+
"grad_norm": 67.16844177246094,
|
44 |
+
"learning_rate": 9.970900000000001e-06,
|
45 |
+
"loss": 2.0949,
|
46 |
+
"step": 300
|
47 |
+
},
|
48 |
+
{
|
49 |
+
"epoch": 1.5384615384615383,
|
50 |
+
"eval_loss": 1.9054155349731445,
|
51 |
+
"eval_runtime": 33.6472,
|
52 |
+
"eval_samples_per_second": 11.68,
|
53 |
+
"eval_steps_per_second": 1.486,
|
54 |
+
"step": 300
|
55 |
+
},
|
56 |
+
{
|
57 |
+
"epoch": 2.051282051282051,
|
58 |
+
"grad_norm": 46.190330505371094,
|
59 |
+
"learning_rate": 9.960900000000001e-06,
|
60 |
+
"loss": 1.9236,
|
61 |
+
"step": 400
|
62 |
+
},
|
63 |
+
{
|
64 |
+
"epoch": 2.051282051282051,
|
65 |
+
"eval_loss": 1.8464823961257935,
|
66 |
+
"eval_runtime": 33.8143,
|
67 |
+
"eval_samples_per_second": 11.622,
|
68 |
+
"eval_steps_per_second": 1.479,
|
69 |
+
"step": 400
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 2.564102564102564,
|
73 |
+
"grad_norm": 42.82326889038086,
|
74 |
+
"learning_rate": 9.950900000000002e-06,
|
75 |
+
"loss": 1.937,
|
76 |
+
"step": 500
|
77 |
+
},
|
78 |
+
{
|
79 |
+
"epoch": 2.564102564102564,
|
80 |
+
"eval_loss": 1.8186373710632324,
|
81 |
+
"eval_runtime": 33.9112,
|
82 |
+
"eval_samples_per_second": 11.589,
|
83 |
+
"eval_steps_per_second": 1.474,
|
84 |
+
"step": 500
|
85 |
+
},
|
86 |
+
{
|
87 |
+
"epoch": 3.076923076923077,
|
88 |
+
"grad_norm": 68.72016906738281,
|
89 |
+
"learning_rate": 9.940900000000002e-06,
|
90 |
+
"loss": 1.8648,
|
91 |
+
"step": 600
|
92 |
+
},
|
93 |
+
{
|
94 |
+
"epoch": 3.076923076923077,
|
95 |
+
"eval_loss": 1.7807129621505737,
|
96 |
+
"eval_runtime": 33.4957,
|
97 |
+
"eval_samples_per_second": 11.733,
|
98 |
+
"eval_steps_per_second": 1.493,
|
99 |
+
"step": 600
|
100 |
+
},
|
101 |
+
{
|
102 |
+
"epoch": 3.58974358974359,
|
103 |
+
"grad_norm": 51.390071868896484,
|
104 |
+
"learning_rate": 9.930900000000002e-06,
|
105 |
+
"loss": 1.8162,
|
106 |
+
"step": 700
|
107 |
+
},
|
108 |
+
{
|
109 |
+
"epoch": 3.58974358974359,
|
110 |
+
"eval_loss": 1.7275961637496948,
|
111 |
+
"eval_runtime": 33.6095,
|
112 |
+
"eval_samples_per_second": 11.693,
|
113 |
+
"eval_steps_per_second": 1.488,
|
114 |
+
"step": 700
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 4.102564102564102,
|
118 |
+
"grad_norm": 33.017948150634766,
|
119 |
+
"learning_rate": 9.920900000000002e-06,
|
120 |
+
"loss": 1.7331,
|
121 |
+
"step": 800
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 4.102564102564102,
|
125 |
+
"eval_loss": 1.701725959777832,
|
126 |
+
"eval_runtime": 33.9602,
|
127 |
+
"eval_samples_per_second": 11.572,
|
128 |
+
"eval_steps_per_second": 1.472,
|
129 |
+
"step": 800
|
130 |
+
},
|
131 |
+
{
|
132 |
+
"epoch": 4.615384615384615,
|
133 |
+
"grad_norm": 43.55423355102539,
|
134 |
+
"learning_rate": 9.9109e-06,
|
135 |
+
"loss": 1.7091,
|
136 |
+
"step": 900
|
137 |
+
},
|
138 |
+
{
|
139 |
+
"epoch": 4.615384615384615,
|
140 |
+
"eval_loss": 1.6944857835769653,
|
141 |
+
"eval_runtime": 33.6785,
|
142 |
+
"eval_samples_per_second": 11.669,
|
143 |
+
"eval_steps_per_second": 1.485,
|
144 |
+
"step": 900
|
145 |
+
},
|
146 |
+
{
|
147 |
+
"epoch": 5.128205128205128,
|
148 |
+
"grad_norm": 67.6618881225586,
|
149 |
+
"learning_rate": 9.9009e-06,
|
150 |
+
"loss": 1.7258,
|
151 |
+
"step": 1000
|
152 |
+
},
|
153 |
+
{
|
154 |
+
"epoch": 5.128205128205128,
|
155 |
+
"eval_loss": 1.65646493434906,
|
156 |
+
"eval_runtime": 33.8316,
|
157 |
+
"eval_samples_per_second": 11.616,
|
158 |
+
"eval_steps_per_second": 1.478,
|
159 |
+
"step": 1000
|
160 |
+
},
|
161 |
+
{
|
162 |
+
"epoch": 5.641025641025641,
|
163 |
+
"grad_norm": 35.863487243652344,
|
164 |
+
"learning_rate": 9.8909e-06,
|
165 |
+
"loss": 1.661,
|
166 |
+
"step": 1100
|
167 |
+
},
|
168 |
+
{
|
169 |
+
"epoch": 5.641025641025641,
|
170 |
+
"eval_loss": 1.6175495386123657,
|
171 |
+
"eval_runtime": 33.57,
|
172 |
+
"eval_samples_per_second": 11.707,
|
173 |
+
"eval_steps_per_second": 1.489,
|
174 |
+
"step": 1100
|
175 |
+
},
|
176 |
+
{
|
177 |
+
"epoch": 6.153846153846154,
|
178 |
+
"grad_norm": 56.98338317871094,
|
179 |
+
"learning_rate": 9.8809e-06,
|
180 |
+
"loss": 1.6368,
|
181 |
+
"step": 1200
|
182 |
+
},
|
183 |
+
{
|
184 |
+
"epoch": 6.153846153846154,
|
185 |
+
"eval_loss": 1.6185818910598755,
|
186 |
+
"eval_runtime": 33.6954,
|
187 |
+
"eval_samples_per_second": 11.663,
|
188 |
+
"eval_steps_per_second": 1.484,
|
189 |
+
"step": 1200
|
190 |
+
},
|
191 |
+
{
|
192 |
+
"epoch": 6.666666666666667,
|
193 |
+
"grad_norm": 90.9850845336914,
|
194 |
+
"learning_rate": 9.8709e-06,
|
195 |
+
"loss": 1.6771,
|
196 |
+
"step": 1300
|
197 |
+
},
|
198 |
+
{
|
199 |
+
"epoch": 6.666666666666667,
|
200 |
+
"eval_loss": 1.597514033317566,
|
201 |
+
"eval_runtime": 33.5635,
|
202 |
+
"eval_samples_per_second": 11.709,
|
203 |
+
"eval_steps_per_second": 1.49,
|
204 |
+
"step": 1300
|
205 |
+
},
|
206 |
+
{
|
207 |
+
"epoch": 7.17948717948718,
|
208 |
+
"grad_norm": 83.6468276977539,
|
209 |
+
"learning_rate": 9.8609e-06,
|
210 |
+
"loss": 1.5796,
|
211 |
+
"step": 1400
|
212 |
+
},
|
213 |
+
{
|
214 |
+
"epoch": 7.17948717948718,
|
215 |
+
"eval_loss": 1.57249116897583,
|
216 |
+
"eval_runtime": 33.6053,
|
217 |
+
"eval_samples_per_second": 11.695,
|
218 |
+
"eval_steps_per_second": 1.488,
|
219 |
+
"step": 1400
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 7.6923076923076925,
|
223 |
+
"grad_norm": 42.52065658569336,
|
224 |
+
"learning_rate": 9.8509e-06,
|
225 |
+
"loss": 1.6012,
|
226 |
+
"step": 1500
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 7.6923076923076925,
|
230 |
+
"eval_loss": 1.5940505266189575,
|
231 |
+
"eval_runtime": 33.6416,
|
232 |
+
"eval_samples_per_second": 11.682,
|
233 |
+
"eval_steps_per_second": 1.486,
|
234 |
+
"step": 1500
|
235 |
+
},
|
236 |
+
{
|
237 |
+
"epoch": 8.205128205128204,
|
238 |
+
"grad_norm": 64.46149444580078,
|
239 |
+
"learning_rate": 9.840900000000001e-06,
|
240 |
+
"loss": 1.6107,
|
241 |
+
"step": 1600
|
242 |
+
},
|
243 |
+
{
|
244 |
+
"epoch": 8.205128205128204,
|
245 |
+
"eval_loss": 1.57711923122406,
|
246 |
+
"eval_runtime": 33.8072,
|
247 |
+
"eval_samples_per_second": 11.625,
|
248 |
+
"eval_steps_per_second": 1.479,
|
249 |
+
"step": 1600
|
250 |
+
},
|
251 |
+
{
|
252 |
+
"epoch": 8.717948717948717,
|
253 |
+
"grad_norm": 47.13843536376953,
|
254 |
+
"learning_rate": 9.830900000000001e-06,
|
255 |
+
"loss": 1.6094,
|
256 |
+
"step": 1700
|
257 |
+
},
|
258 |
+
{
|
259 |
+
"epoch": 8.717948717948717,
|
260 |
+
"eval_loss": 1.5567021369934082,
|
261 |
+
"eval_runtime": 33.6256,
|
262 |
+
"eval_samples_per_second": 11.688,
|
263 |
+
"eval_steps_per_second": 1.487,
|
264 |
+
"step": 1700
|
265 |
+
},
|
266 |
+
{
|
267 |
+
"epoch": 9.23076923076923,
|
268 |
+
"grad_norm": 60.96649932861328,
|
269 |
+
"learning_rate": 9.820900000000001e-06,
|
270 |
+
"loss": 1.5679,
|
271 |
+
"step": 1800
|
272 |
+
},
|
273 |
+
{
|
274 |
+
"epoch": 9.23076923076923,
|
275 |
+
"eval_loss": 1.5404459238052368,
|
276 |
+
"eval_runtime": 33.5651,
|
277 |
+
"eval_samples_per_second": 11.709,
|
278 |
+
"eval_steps_per_second": 1.49,
|
279 |
+
"step": 1800
|
280 |
+
},
|
281 |
+
{
|
282 |
+
"epoch": 9.743589743589745,
|
283 |
+
"grad_norm": 48.0488395690918,
|
284 |
+
"learning_rate": 9.810900000000001e-06,
|
285 |
+
"loss": 1.541,
|
286 |
+
"step": 1900
|
287 |
+
},
|
288 |
+
{
|
289 |
+
"epoch": 9.743589743589745,
|
290 |
+
"eval_loss": 1.5000860691070557,
|
291 |
+
"eval_runtime": 33.729,
|
292 |
+
"eval_samples_per_second": 11.652,
|
293 |
+
"eval_steps_per_second": 1.482,
|
294 |
+
"step": 1900
|
295 |
+
},
|
296 |
+
{
|
297 |
+
"epoch": 10.256410256410255,
|
298 |
+
"grad_norm": 21.072124481201172,
|
299 |
+
"learning_rate": 9.800900000000001e-06,
|
300 |
+
"loss": 1.4747,
|
301 |
+
"step": 2000
|
302 |
+
},
|
303 |
+
{
|
304 |
+
"epoch": 10.256410256410255,
|
305 |
+
"eval_loss": 1.4845048189163208,
|
306 |
+
"eval_runtime": 33.6789,
|
307 |
+
"eval_samples_per_second": 11.669,
|
308 |
+
"eval_steps_per_second": 1.485,
|
309 |
+
"step": 2000
|
310 |
+
},
|
311 |
+
{
|
312 |
+
"epoch": 10.76923076923077,
|
313 |
+
"grad_norm": 60.55208969116211,
|
314 |
+
"learning_rate": 9.790900000000001e-06,
|
315 |
+
"loss": 1.5105,
|
316 |
+
"step": 2100
|
317 |
+
},
|
318 |
+
{
|
319 |
+
"epoch": 10.76923076923077,
|
320 |
+
"eval_loss": 1.4587421417236328,
|
321 |
+
"eval_runtime": 33.4313,
|
322 |
+
"eval_samples_per_second": 11.755,
|
323 |
+
"eval_steps_per_second": 1.496,
|
324 |
+
"step": 2100
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 11.282051282051283,
|
328 |
+
"grad_norm": 26.457704544067383,
|
329 |
+
"learning_rate": 9.780900000000002e-06,
|
330 |
+
"loss": 1.4769,
|
331 |
+
"step": 2200
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 11.282051282051283,
|
335 |
+
"eval_loss": 1.4588648080825806,
|
336 |
+
"eval_runtime": 33.5978,
|
337 |
+
"eval_samples_per_second": 11.697,
|
338 |
+
"eval_steps_per_second": 1.488,
|
339 |
+
"step": 2200
|
340 |
+
},
|
341 |
+
{
|
342 |
+
"epoch": 11.794871794871796,
|
343 |
+
"grad_norm": 23.801916122436523,
|
344 |
+
"learning_rate": 9.770900000000002e-06,
|
345 |
+
"loss": 1.4555,
|
346 |
+
"step": 2300
|
347 |
+
},
|
348 |
+
{
|
349 |
+
"epoch": 11.794871794871796,
|
350 |
+
"eval_loss": 1.4805899858474731,
|
351 |
+
"eval_runtime": 33.4947,
|
352 |
+
"eval_samples_per_second": 11.733,
|
353 |
+
"eval_steps_per_second": 1.493,
|
354 |
+
"step": 2300
|
355 |
+
},
|
356 |
+
{
|
357 |
+
"epoch": 12.307692307692308,
|
358 |
+
"grad_norm": 24.780275344848633,
|
359 |
+
"learning_rate": 9.760900000000002e-06,
|
360 |
+
"loss": 1.4537,
|
361 |
+
"step": 2400
|
362 |
+
},
|
363 |
+
{
|
364 |
+
"epoch": 12.307692307692308,
|
365 |
+
"eval_loss": 1.4648451805114746,
|
366 |
+
"eval_runtime": 33.6877,
|
367 |
+
"eval_samples_per_second": 11.666,
|
368 |
+
"eval_steps_per_second": 1.484,
|
369 |
+
"step": 2400
|
370 |
+
},
|
371 |
+
{
|
372 |
+
"epoch": 12.820512820512821,
|
373 |
+
"grad_norm": 65.28894805908203,
|
374 |
+
"learning_rate": 9.7509e-06,
|
375 |
+
"loss": 1.4788,
|
376 |
+
"step": 2500
|
377 |
+
},
|
378 |
+
{
|
379 |
+
"epoch": 12.820512820512821,
|
380 |
+
"eval_loss": 1.4453215599060059,
|
381 |
+
"eval_runtime": 33.972,
|
382 |
+
"eval_samples_per_second": 11.568,
|
383 |
+
"eval_steps_per_second": 1.472,
|
384 |
+
"step": 2500
|
385 |
+
},
|
386 |
+
{
|
387 |
+
"epoch": 13.333333333333334,
|
388 |
+
"grad_norm": 23.36334800720215,
|
389 |
+
"learning_rate": 9.7409e-06,
|
390 |
+
"loss": 1.4397,
|
391 |
+
"step": 2600
|
392 |
+
},
|
393 |
+
{
|
394 |
+
"epoch": 13.333333333333334,
|
395 |
+
"eval_loss": 1.4110159873962402,
|
396 |
+
"eval_runtime": 33.3905,
|
397 |
+
"eval_samples_per_second": 11.77,
|
398 |
+
"eval_steps_per_second": 1.497,
|
399 |
+
"step": 2600
|
400 |
+
},
|
401 |
+
{
|
402 |
+
"epoch": 13.846153846153847,
|
403 |
+
"grad_norm": 24.13091468811035,
|
404 |
+
"learning_rate": 9.7309e-06,
|
405 |
+
"loss": 1.42,
|
406 |
+
"step": 2700
|
407 |
+
},
|
408 |
+
{
|
409 |
+
"epoch": 13.846153846153847,
|
410 |
+
"eval_loss": 1.3875733613967896,
|
411 |
+
"eval_runtime": 33.5662,
|
412 |
+
"eval_samples_per_second": 11.708,
|
413 |
+
"eval_steps_per_second": 1.49,
|
414 |
+
"step": 2700
|
415 |
+
},
|
416 |
+
{
|
417 |
+
"epoch": 14.35897435897436,
|
418 |
+
"grad_norm": 41.02390670776367,
|
419 |
+
"learning_rate": 9.7209e-06,
|
420 |
+
"loss": 1.3833,
|
421 |
+
"step": 2800
|
422 |
+
},
|
423 |
+
{
|
424 |
+
"epoch": 14.35897435897436,
|
425 |
+
"eval_loss": 1.3958121538162231,
|
426 |
+
"eval_runtime": 33.6111,
|
427 |
+
"eval_samples_per_second": 11.693,
|
428 |
+
"eval_steps_per_second": 1.488,
|
429 |
+
"step": 2800
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 14.871794871794872,
|
433 |
+
"grad_norm": 25.11317253112793,
|
434 |
+
"learning_rate": 9.7109e-06,
|
435 |
+
"loss": 1.3839,
|
436 |
+
"step": 2900
|
437 |
+
},
|
438 |
+
{
|
439 |
+
"epoch": 14.871794871794872,
|
440 |
+
"eval_loss": 1.4024689197540283,
|
441 |
+
"eval_runtime": 33.5271,
|
442 |
+
"eval_samples_per_second": 11.722,
|
443 |
+
"eval_steps_per_second": 1.491,
|
444 |
+
"step": 2900
|
445 |
+
},
|
446 |
+
{
|
447 |
+
"epoch": 15.384615384615385,
|
448 |
+
"grad_norm": 32.921974182128906,
|
449 |
+
"learning_rate": 9.7009e-06,
|
450 |
+
"loss": 1.3998,
|
451 |
+
"step": 3000
|
452 |
+
},
|
453 |
+
{
|
454 |
+
"epoch": 15.384615384615385,
|
455 |
+
"eval_loss": 1.3834290504455566,
|
456 |
+
"eval_runtime": 33.4449,
|
457 |
+
"eval_samples_per_second": 11.751,
|
458 |
+
"eval_steps_per_second": 1.495,
|
459 |
+
"step": 3000
|
460 |
+
},
|
461 |
+
{
|
462 |
+
"epoch": 15.897435897435898,
|
463 |
+
"grad_norm": 71.14076232910156,
|
464 |
+
"learning_rate": 9.6909e-06,
|
465 |
+
"loss": 1.3753,
|
466 |
+
"step": 3100
|
467 |
+
},
|
468 |
+
{
|
469 |
+
"epoch": 15.897435897435898,
|
470 |
+
"eval_loss": 1.371915578842163,
|
471 |
+
"eval_runtime": 33.5787,
|
472 |
+
"eval_samples_per_second": 11.704,
|
473 |
+
"eval_steps_per_second": 1.489,
|
474 |
+
"step": 3100
|
475 |
+
},
|
476 |
+
{
|
477 |
+
"epoch": 16.41025641025641,
|
478 |
+
"grad_norm": 85.71231842041016,
|
479 |
+
"learning_rate": 9.6809e-06,
|
480 |
+
"loss": 1.3639,
|
481 |
+
"step": 3200
|
482 |
+
},
|
483 |
+
{
|
484 |
+
"epoch": 16.41025641025641,
|
485 |
+
"eval_loss": 1.3810721635818481,
|
486 |
+
"eval_runtime": 33.5802,
|
487 |
+
"eval_samples_per_second": 11.703,
|
488 |
+
"eval_steps_per_second": 1.489,
|
489 |
+
"step": 3200
|
490 |
+
},
|
491 |
+
{
|
492 |
+
"epoch": 16.923076923076923,
|
493 |
+
"grad_norm": 39.7967529296875,
|
494 |
+
"learning_rate": 9.670900000000001e-06,
|
495 |
+
"loss": 1.3609,
|
496 |
+
"step": 3300
|
497 |
+
},
|
498 |
+
{
|
499 |
+
"epoch": 16.923076923076923,
|
500 |
+
"eval_loss": 1.3853365182876587,
|
501 |
+
"eval_runtime": 33.6581,
|
502 |
+
"eval_samples_per_second": 11.676,
|
503 |
+
"eval_steps_per_second": 1.486,
|
504 |
+
"step": 3300
|
505 |
+
},
|
506 |
+
{
|
507 |
+
"epoch": 17.435897435897434,
|
508 |
+
"grad_norm": 19.84697723388672,
|
509 |
+
"learning_rate": 9.660900000000001e-06,
|
510 |
+
"loss": 1.3314,
|
511 |
+
"step": 3400
|
512 |
+
},
|
513 |
+
{
|
514 |
+
"epoch": 17.435897435897434,
|
515 |
+
"eval_loss": 1.3581265211105347,
|
516 |
+
"eval_runtime": 33.4833,
|
517 |
+
"eval_samples_per_second": 11.737,
|
518 |
+
"eval_steps_per_second": 1.493,
|
519 |
+
"step": 3400
|
520 |
+
},
|
521 |
+
{
|
522 |
+
"epoch": 17.94871794871795,
|
523 |
+
"grad_norm": 43.280426025390625,
|
524 |
+
"learning_rate": 9.650900000000001e-06,
|
525 |
+
"loss": 1.3408,
|
526 |
+
"step": 3500
|
527 |
+
},
|
528 |
+
{
|
529 |
+
"epoch": 17.94871794871795,
|
530 |
+
"eval_loss": 1.3400689363479614,
|
531 |
+
"eval_runtime": 33.4394,
|
532 |
+
"eval_samples_per_second": 11.753,
|
533 |
+
"eval_steps_per_second": 1.495,
|
534 |
+
"step": 3500
|
535 |
+
},
|
536 |
+
{
|
537 |
+
"epoch": 18.46153846153846,
|
538 |
+
"grad_norm": 35.393314361572266,
|
539 |
+
"learning_rate": 9.640900000000001e-06,
|
540 |
+
"loss": 1.3471,
|
541 |
+
"step": 3600
|
542 |
+
},
|
543 |
+
{
|
544 |
+
"epoch": 18.46153846153846,
|
545 |
+
"eval_loss": 1.3405698537826538,
|
546 |
+
"eval_runtime": 33.6347,
|
547 |
+
"eval_samples_per_second": 11.684,
|
548 |
+
"eval_steps_per_second": 1.487,
|
549 |
+
"step": 3600
|
550 |
+
},
|
551 |
+
{
|
552 |
+
"epoch": 18.974358974358974,
|
553 |
+
"grad_norm": 32.50804138183594,
|
554 |
+
"learning_rate": 9.630900000000001e-06,
|
555 |
+
"loss": 1.3213,
|
556 |
+
"step": 3700
|
557 |
+
},
|
558 |
+
{
|
559 |
+
"epoch": 18.974358974358974,
|
560 |
+
"eval_loss": 1.3345987796783447,
|
561 |
+
"eval_runtime": 33.5369,
|
562 |
+
"eval_samples_per_second": 11.718,
|
563 |
+
"eval_steps_per_second": 1.491,
|
564 |
+
"step": 3700
|
565 |
+
},
|
566 |
+
{
|
567 |
+
"epoch": 19.487179487179485,
|
568 |
+
"grad_norm": 28.999061584472656,
|
569 |
+
"learning_rate": 9.620900000000001e-06,
|
570 |
+
"loss": 1.3251,
|
571 |
+
"step": 3800
|
572 |
+
},
|
573 |
+
{
|
574 |
+
"epoch": 19.487179487179485,
|
575 |
+
"eval_loss": 1.3227241039276123,
|
576 |
+
"eval_runtime": 33.5001,
|
577 |
+
"eval_samples_per_second": 11.731,
|
578 |
+
"eval_steps_per_second": 1.493,
|
579 |
+
"step": 3800
|
580 |
+
},
|
581 |
+
{
|
582 |
+
"epoch": 20.0,
|
583 |
+
"grad_norm": 41.793636322021484,
|
584 |
+
"learning_rate": 9.610900000000001e-06,
|
585 |
+
"loss": 1.3297,
|
586 |
+
"step": 3900
|
587 |
+
},
|
588 |
+
{
|
589 |
+
"epoch": 20.0,
|
590 |
+
"eval_loss": 1.3213008642196655,
|
591 |
+
"eval_runtime": 33.561,
|
592 |
+
"eval_samples_per_second": 11.71,
|
593 |
+
"eval_steps_per_second": 1.49,
|
594 |
+
"step": 3900
|
595 |
+
},
|
596 |
+
{
|
597 |
+
"epoch": 20.51282051282051,
|
598 |
+
"grad_norm": 53.55360794067383,
|
599 |
+
"learning_rate": 9.600900000000002e-06,
|
600 |
+
"loss": 1.331,
|
601 |
+
"step": 4000
|
602 |
+
},
|
603 |
+
{
|
604 |
+
"epoch": 20.51282051282051,
|
605 |
+
"eval_loss": 1.3293949365615845,
|
606 |
+
"eval_runtime": 33.5542,
|
607 |
+
"eval_samples_per_second": 11.712,
|
608 |
+
"eval_steps_per_second": 1.49,
|
609 |
+
"step": 4000
|
610 |
+
},
|
611 |
+
{
|
612 |
+
"epoch": 21.025641025641026,
|
613 |
+
"grad_norm": 73.3389892578125,
|
614 |
+
"learning_rate": 9.5909e-06,
|
615 |
+
"loss": 1.2856,
|
616 |
+
"step": 4100
|
617 |
+
},
|
618 |
+
{
|
619 |
+
"epoch": 21.025641025641026,
|
620 |
+
"eval_loss": 1.316487193107605,
|
621 |
+
"eval_runtime": 33.5602,
|
622 |
+
"eval_samples_per_second": 11.71,
|
623 |
+
"eval_steps_per_second": 1.49,
|
624 |
+
"step": 4100
|
625 |
+
},
|
626 |
+
{
|
627 |
+
"epoch": 21.53846153846154,
|
628 |
+
"grad_norm": 62.724456787109375,
|
629 |
+
"learning_rate": 9.5809e-06,
|
630 |
+
"loss": 1.3186,
|
631 |
+
"step": 4200
|
632 |
+
},
|
633 |
+
{
|
634 |
+
"epoch": 21.53846153846154,
|
635 |
+
"eval_loss": 1.3179290294647217,
|
636 |
+
"eval_runtime": 33.621,
|
637 |
+
"eval_samples_per_second": 11.689,
|
638 |
+
"eval_steps_per_second": 1.487,
|
639 |
+
"step": 4200
|
640 |
+
},
|
641 |
+
{
|
642 |
+
"epoch": 22.05128205128205,
|
643 |
+
"grad_norm": 30.242937088012695,
|
644 |
+
"learning_rate": 9.5709e-06,
|
645 |
+
"loss": 1.2623,
|
646 |
+
"step": 4300
|
647 |
+
},
|
648 |
+
{
|
649 |
+
"epoch": 22.05128205128205,
|
650 |
+
"eval_loss": 1.2920558452606201,
|
651 |
+
"eval_runtime": 33.6166,
|
652 |
+
"eval_samples_per_second": 11.691,
|
653 |
+
"eval_steps_per_second": 1.487,
|
654 |
+
"step": 4300
|
655 |
+
},
|
656 |
+
{
|
657 |
+
"epoch": 22.564102564102566,
|
658 |
+
"grad_norm": 17.423118591308594,
|
659 |
+
"learning_rate": 9.5609e-06,
|
660 |
+
"loss": 1.3045,
|
661 |
+
"step": 4400
|
662 |
+
},
|
663 |
+
{
|
664 |
+
"epoch": 22.564102564102566,
|
665 |
+
"eval_loss": 1.302635669708252,
|
666 |
+
"eval_runtime": 33.7322,
|
667 |
+
"eval_samples_per_second": 11.651,
|
668 |
+
"eval_steps_per_second": 1.482,
|
669 |
+
"step": 4400
|
670 |
+
},
|
671 |
+
{
|
672 |
+
"epoch": 23.076923076923077,
|
673 |
+
"grad_norm": 115.05293273925781,
|
674 |
+
"learning_rate": 9.5509e-06,
|
675 |
+
"loss": 1.2997,
|
676 |
+
"step": 4500
|
677 |
+
},
|
678 |
+
{
|
679 |
+
"epoch": 23.076923076923077,
|
680 |
+
"eval_loss": 1.2834452390670776,
|
681 |
+
"eval_runtime": 33.5349,
|
682 |
+
"eval_samples_per_second": 11.719,
|
683 |
+
"eval_steps_per_second": 1.491,
|
684 |
+
"step": 4500
|
685 |
+
},
|
686 |
+
{
|
687 |
+
"epoch": 23.58974358974359,
|
688 |
+
"grad_norm": 35.13385009765625,
|
689 |
+
"learning_rate": 9.5409e-06,
|
690 |
+
"loss": 1.2591,
|
691 |
+
"step": 4600
|
692 |
+
},
|
693 |
+
{
|
694 |
+
"epoch": 23.58974358974359,
|
695 |
+
"eval_loss": 1.2677439451217651,
|
696 |
+
"eval_runtime": 33.5991,
|
697 |
+
"eval_samples_per_second": 11.697,
|
698 |
+
"eval_steps_per_second": 1.488,
|
699 |
+
"step": 4600
|
700 |
+
},
|
701 |
+
{
|
702 |
+
"epoch": 24.102564102564102,
|
703 |
+
"grad_norm": 22.429704666137695,
|
704 |
+
"learning_rate": 9.5309e-06,
|
705 |
+
"loss": 1.3057,
|
706 |
+
"step": 4700
|
707 |
+
},
|
708 |
+
{
|
709 |
+
"epoch": 24.102564102564102,
|
710 |
+
"eval_loss": 1.2759162187576294,
|
711 |
+
"eval_runtime": 33.8336,
|
712 |
+
"eval_samples_per_second": 11.616,
|
713 |
+
"eval_steps_per_second": 1.478,
|
714 |
+
"step": 4700
|
715 |
+
},
|
716 |
+
{
|
717 |
+
"epoch": 24.615384615384617,
|
718 |
+
"grad_norm": 19.258209228515625,
|
719 |
+
"learning_rate": 9.5209e-06,
|
720 |
+
"loss": 1.2526,
|
721 |
+
"step": 4800
|
722 |
+
},
|
723 |
+
{
|
724 |
+
"epoch": 24.615384615384617,
|
725 |
+
"eval_loss": 1.2712483406066895,
|
726 |
+
"eval_runtime": 33.585,
|
727 |
+
"eval_samples_per_second": 11.702,
|
728 |
+
"eval_steps_per_second": 1.489,
|
729 |
+
"step": 4800
|
730 |
+
},
|
731 |
+
{
|
732 |
+
"epoch": 25.128205128205128,
|
733 |
+
"grad_norm": 18.289043426513672,
|
734 |
+
"learning_rate": 9.5109e-06,
|
735 |
+
"loss": 1.2436,
|
736 |
+
"step": 4900
|
737 |
+
},
|
738 |
+
{
|
739 |
+
"epoch": 25.128205128205128,
|
740 |
+
"eval_loss": 1.277999758720398,
|
741 |
+
"eval_runtime": 33.6639,
|
742 |
+
"eval_samples_per_second": 11.674,
|
743 |
+
"eval_steps_per_second": 1.485,
|
744 |
+
"step": 4900
|
745 |
+
},
|
746 |
+
{
|
747 |
+
"epoch": 25.641025641025642,
|
748 |
+
"grad_norm": 53.41452407836914,
|
749 |
+
"learning_rate": 9.501000000000001e-06,
|
750 |
+
"loss": 1.2768,
|
751 |
+
"step": 5000
|
752 |
+
},
|
753 |
+
{
|
754 |
+
"epoch": 25.641025641025642,
|
755 |
+
"eval_loss": 1.2602609395980835,
|
756 |
+
"eval_runtime": 33.6751,
|
757 |
+
"eval_samples_per_second": 11.67,
|
758 |
+
"eval_steps_per_second": 1.485,
|
759 |
+
"step": 5000
|
760 |
+
},
|
761 |
+
{
|
762 |
+
"epoch": 26.153846153846153,
|
763 |
+
"grad_norm": 28.17922019958496,
|
764 |
+
"learning_rate": 9.491000000000001e-06,
|
765 |
+
"loss": 1.2221,
|
766 |
+
"step": 5100
|
767 |
+
},
|
768 |
+
{
|
769 |
+
"epoch": 26.153846153846153,
|
770 |
+
"eval_loss": 1.28999662399292,
|
771 |
+
"eval_runtime": 33.8104,
|
772 |
+
"eval_samples_per_second": 11.624,
|
773 |
+
"eval_steps_per_second": 1.479,
|
774 |
+
"step": 5100
|
775 |
+
},
|
776 |
+
{
|
777 |
+
"epoch": 26.666666666666668,
|
778 |
+
"grad_norm": 28.119773864746094,
|
779 |
+
"learning_rate": 9.481000000000001e-06,
|
780 |
+
"loss": 1.233,
|
781 |
+
"step": 5200
|
782 |
+
},
|
783 |
+
{
|
784 |
+
"epoch": 26.666666666666668,
|
785 |
+
"eval_loss": 1.2724733352661133,
|
786 |
+
"eval_runtime": 33.9553,
|
787 |
+
"eval_samples_per_second": 11.574,
|
788 |
+
"eval_steps_per_second": 1.473,
|
789 |
+
"step": 5200
|
790 |
+
},
|
791 |
+
{
|
792 |
+
"epoch": 27.17948717948718,
|
793 |
+
"grad_norm": 102.97128295898438,
|
794 |
+
"learning_rate": 9.471000000000001e-06,
|
795 |
+
"loss": 1.2667,
|
796 |
+
"step": 5300
|
797 |
+
},
|
798 |
+
{
|
799 |
+
"epoch": 27.17948717948718,
|
800 |
+
"eval_loss": 1.265009880065918,
|
801 |
+
"eval_runtime": 33.7879,
|
802 |
+
"eval_samples_per_second": 11.631,
|
803 |
+
"eval_steps_per_second": 1.48,
|
804 |
+
"step": 5300
|
805 |
+
},
|
806 |
+
{
|
807 |
+
"epoch": 27.692307692307693,
|
808 |
+
"grad_norm": 23.439664840698242,
|
809 |
+
"learning_rate": 9.461000000000001e-06,
|
810 |
+
"loss": 1.2255,
|
811 |
+
"step": 5400
|
812 |
+
},
|
813 |
+
{
|
814 |
+
"epoch": 27.692307692307693,
|
815 |
+
"eval_loss": 1.2421387434005737,
|
816 |
+
"eval_runtime": 33.867,
|
817 |
+
"eval_samples_per_second": 11.604,
|
818 |
+
"eval_steps_per_second": 1.476,
|
819 |
+
"step": 5400
|
820 |
+
},
|
821 |
+
{
|
822 |
+
"epoch": 28.205128205128204,
|
823 |
+
"grad_norm": 34.7141227722168,
|
824 |
+
"learning_rate": 9.451000000000002e-06,
|
825 |
+
"loss": 1.2557,
|
826 |
+
"step": 5500
|
827 |
+
},
|
828 |
+
{
|
829 |
+
"epoch": 28.205128205128204,
|
830 |
+
"eval_loss": 1.2613080739974976,
|
831 |
+
"eval_runtime": 33.9538,
|
832 |
+
"eval_samples_per_second": 11.575,
|
833 |
+
"eval_steps_per_second": 1.473,
|
834 |
+
"step": 5500
|
835 |
+
},
|
836 |
+
{
|
837 |
+
"epoch": 28.71794871794872,
|
838 |
+
"grad_norm": 60.65612030029297,
|
839 |
+
"learning_rate": 9.441000000000002e-06,
|
840 |
+
"loss": 1.2407,
|
841 |
+
"step": 5600
|
842 |
+
},
|
843 |
+
{
|
844 |
+
"epoch": 28.71794871794872,
|
845 |
+
"eval_loss": 1.2578727006912231,
|
846 |
+
"eval_runtime": 33.8423,
|
847 |
+
"eval_samples_per_second": 11.613,
|
848 |
+
"eval_steps_per_second": 1.477,
|
849 |
+
"step": 5600
|
850 |
+
},
|
851 |
+
{
|
852 |
+
"epoch": 29.23076923076923,
|
853 |
+
"grad_norm": 39.73835754394531,
|
854 |
+
"learning_rate": 9.431000000000002e-06,
|
855 |
+
"loss": 1.2436,
|
856 |
+
"step": 5700
|
857 |
+
},
|
858 |
+
{
|
859 |
+
"epoch": 29.23076923076923,
|
860 |
+
"eval_loss": 1.2593867778778076,
|
861 |
+
"eval_runtime": 33.6929,
|
862 |
+
"eval_samples_per_second": 11.664,
|
863 |
+
"eval_steps_per_second": 1.484,
|
864 |
+
"step": 5700
|
865 |
+
},
|
866 |
+
{
|
867 |
+
"epoch": 29.743589743589745,
|
868 |
+
"grad_norm": 24.647741317749023,
|
869 |
+
"learning_rate": 9.421000000000002e-06,
|
870 |
+
"loss": 1.207,
|
871 |
+
"step": 5800
|
872 |
+
},
|
873 |
+
{
|
874 |
+
"epoch": 29.743589743589745,
|
875 |
+
"eval_loss": 1.2388144731521606,
|
876 |
+
"eval_runtime": 33.7238,
|
877 |
+
"eval_samples_per_second": 11.654,
|
878 |
+
"eval_steps_per_second": 1.483,
|
879 |
+
"step": 5800
|
880 |
+
},
|
881 |
+
{
|
882 |
+
"epoch": 30.256410256410255,
|
883 |
+
"grad_norm": 63.13822937011719,
|
884 |
+
"learning_rate": 9.411000000000002e-06,
|
885 |
+
"loss": 1.2287,
|
886 |
+
"step": 5900
|
887 |
+
},
|
888 |
+
{
|
889 |
+
"epoch": 30.256410256410255,
|
890 |
+
"eval_loss": 1.2804478406906128,
|
891 |
+
"eval_runtime": 33.7389,
|
892 |
+
"eval_samples_per_second": 11.648,
|
893 |
+
"eval_steps_per_second": 1.482,
|
894 |
+
"step": 5900
|
895 |
+
},
|
896 |
+
{
|
897 |
+
"epoch": 30.76923076923077,
|
898 |
+
"grad_norm": 27.123302459716797,
|
899 |
+
"learning_rate": 9.401000000000002e-06,
|
900 |
+
"loss": 1.2249,
|
901 |
+
"step": 6000
|
902 |
+
},
|
903 |
+
{
|
904 |
+
"epoch": 30.76923076923077,
|
905 |
+
"eval_loss": 1.2437323331832886,
|
906 |
+
"eval_runtime": 33.6856,
|
907 |
+
"eval_samples_per_second": 11.667,
|
908 |
+
"eval_steps_per_second": 1.484,
|
909 |
+
"step": 6000
|
910 |
+
},
|
911 |
+
{
|
912 |
+
"epoch": 31.28205128205128,
|
913 |
+
"grad_norm": 27.28360366821289,
|
914 |
+
"learning_rate": 9.391e-06,
|
915 |
+
"loss": 1.2101,
|
916 |
+
"step": 6100
|
917 |
+
},
|
918 |
+
{
|
919 |
+
"epoch": 31.28205128205128,
|
920 |
+
"eval_loss": 1.227403998374939,
|
921 |
+
"eval_runtime": 33.648,
|
922 |
+
"eval_samples_per_second": 11.68,
|
923 |
+
"eval_steps_per_second": 1.486,
|
924 |
+
"step": 6100
|
925 |
+
},
|
926 |
+
{
|
927 |
+
"epoch": 31.794871794871796,
|
928 |
+
"grad_norm": 55.08905029296875,
|
929 |
+
"learning_rate": 9.381e-06,
|
930 |
+
"loss": 1.1843,
|
931 |
+
"step": 6200
|
932 |
+
},
|
933 |
+
{
|
934 |
+
"epoch": 31.794871794871796,
|
935 |
+
"eval_loss": 1.225257158279419,
|
936 |
+
"eval_runtime": 33.8349,
|
937 |
+
"eval_samples_per_second": 11.615,
|
938 |
+
"eval_steps_per_second": 1.478,
|
939 |
+
"step": 6200
|
940 |
+
},
|
941 |
+
{
|
942 |
+
"epoch": 32.30769230769231,
|
943 |
+
"grad_norm": 39.94312286376953,
|
944 |
+
"learning_rate": 9.371e-06,
|
945 |
+
"loss": 1.1908,
|
946 |
+
"step": 6300
|
947 |
+
},
|
948 |
+
{
|
949 |
+
"epoch": 32.30769230769231,
|
950 |
+
"eval_loss": 1.2344192266464233,
|
951 |
+
"eval_runtime": 33.9417,
|
952 |
+
"eval_samples_per_second": 11.579,
|
953 |
+
"eval_steps_per_second": 1.473,
|
954 |
+
"step": 6300
|
955 |
+
},
|
956 |
+
{
|
957 |
+
"epoch": 32.82051282051282,
|
958 |
+
"grad_norm": 25.61664581298828,
|
959 |
+
"learning_rate": 9.361e-06,
|
960 |
+
"loss": 1.1832,
|
961 |
+
"step": 6400
|
962 |
+
},
|
963 |
+
{
|
964 |
+
"epoch": 32.82051282051282,
|
965 |
+
"eval_loss": 1.2213290929794312,
|
966 |
+
"eval_runtime": 33.8461,
|
967 |
+
"eval_samples_per_second": 11.611,
|
968 |
+
"eval_steps_per_second": 1.477,
|
969 |
+
"step": 6400
|
970 |
+
},
|
971 |
+
{
|
972 |
+
"epoch": 33.333333333333336,
|
973 |
+
"grad_norm": 31.0895938873291,
|
974 |
+
"learning_rate": 9.351e-06,
|
975 |
+
"loss": 1.2104,
|
976 |
+
"step": 6500
|
977 |
+
},
|
978 |
+
{
|
979 |
+
"epoch": 33.333333333333336,
|
980 |
+
"eval_loss": 1.2290922403335571,
|
981 |
+
"eval_runtime": 33.8119,
|
982 |
+
"eval_samples_per_second": 11.623,
|
983 |
+
"eval_steps_per_second": 1.479,
|
984 |
+
"step": 6500
|
985 |
+
},
|
986 |
+
{
|
987 |
+
"epoch": 33.84615384615385,
|
988 |
+
"grad_norm": 35.208988189697266,
|
989 |
+
"learning_rate": 9.341000000000001e-06,
|
990 |
+
"loss": 1.2007,
|
991 |
+
"step": 6600
|
992 |
+
},
|
993 |
+
{
|
994 |
+
"epoch": 33.84615384615385,
|
995 |
+
"eval_loss": 1.2198638916015625,
|
996 |
+
"eval_runtime": 33.6979,
|
997 |
+
"eval_samples_per_second": 11.662,
|
998 |
+
"eval_steps_per_second": 1.484,
|
999 |
+
"step": 6600
|
1000 |
+
},
|
1001 |
+
{
|
1002 |
+
"epoch": 34.35897435897436,
|
1003 |
+
"grad_norm": 46.61259841918945,
|
1004 |
+
"learning_rate": 9.331000000000001e-06,
|
1005 |
+
"loss": 1.182,
|
1006 |
+
"step": 6700
|
1007 |
+
},
|
1008 |
+
{
|
1009 |
+
"epoch": 34.35897435897436,
|
1010 |
+
"eval_loss": 1.202943205833435,
|
1011 |
+
"eval_runtime": 33.8788,
|
1012 |
+
"eval_samples_per_second": 11.6,
|
1013 |
+
"eval_steps_per_second": 1.476,
|
1014 |
+
"step": 6700
|
1015 |
+
},
|
1016 |
+
{
|
1017 |
+
"epoch": 34.87179487179487,
|
1018 |
+
"grad_norm": 27.606361389160156,
|
1019 |
+
"learning_rate": 9.321000000000001e-06,
|
1020 |
+
"loss": 1.1843,
|
1021 |
+
"step": 6800
|
1022 |
+
},
|
1023 |
+
{
|
1024 |
+
"epoch": 34.87179487179487,
|
1025 |
+
"eval_loss": 1.2101120948791504,
|
1026 |
+
"eval_runtime": 33.92,
|
1027 |
+
"eval_samples_per_second": 11.586,
|
1028 |
+
"eval_steps_per_second": 1.474,
|
1029 |
+
"step": 6800
|
1030 |
+
},
|
1031 |
+
{
|
1032 |
+
"epoch": 35.38461538461539,
|
1033 |
+
"grad_norm": 1141.704833984375,
|
1034 |
+
"learning_rate": 9.311000000000001e-06,
|
1035 |
+
"loss": 1.1926,
|
1036 |
+
"step": 6900
|
1037 |
+
},
|
1038 |
+
{
|
1039 |
+
"epoch": 35.38461538461539,
|
1040 |
+
"eval_loss": 1.1979094743728638,
|
1041 |
+
"eval_runtime": 33.9794,
|
1042 |
+
"eval_samples_per_second": 11.566,
|
1043 |
+
"eval_steps_per_second": 1.471,
|
1044 |
+
"step": 6900
|
1045 |
+
},
|
1046 |
+
{
|
1047 |
+
"epoch": 35.8974358974359,
|
1048 |
+
"grad_norm": 29.9624080657959,
|
1049 |
+
"learning_rate": 9.301000000000001e-06,
|
1050 |
+
"loss": 1.1475,
|
1051 |
+
"step": 7000
|
1052 |
+
},
|
1053 |
+
{
|
1054 |
+
"epoch": 35.8974358974359,
|
1055 |
+
"eval_loss": 1.1987537145614624,
|
1056 |
+
"eval_runtime": 33.7554,
|
1057 |
+
"eval_samples_per_second": 11.643,
|
1058 |
+
"eval_steps_per_second": 1.481,
|
1059 |
+
"step": 7000
|
1060 |
+
},
|
1061 |
+
{
|
1062 |
+
"epoch": 36.41025641025641,
|
1063 |
+
"grad_norm": 23.925745010375977,
|
1064 |
+
"learning_rate": 9.291000000000001e-06,
|
1065 |
+
"loss": 1.1692,
|
1066 |
+
"step": 7100
|
1067 |
+
},
|
1068 |
+
{
|
1069 |
+
"epoch": 36.41025641025641,
|
1070 |
+
"eval_loss": 1.1971298456192017,
|
1071 |
+
"eval_runtime": 33.682,
|
1072 |
+
"eval_samples_per_second": 11.668,
|
1073 |
+
"eval_steps_per_second": 1.484,
|
1074 |
+
"step": 7100
|
1075 |
+
},
|
1076 |
+
{
|
1077 |
+
"epoch": 36.92307692307692,
|
1078 |
+
"grad_norm": 22.790334701538086,
|
1079 |
+
"learning_rate": 9.281000000000001e-06,
|
1080 |
+
"loss": 1.1381,
|
1081 |
+
"step": 7200
|
1082 |
+
},
|
1083 |
+
{
|
1084 |
+
"epoch": 36.92307692307692,
|
1085 |
+
"eval_loss": 1.1832250356674194,
|
1086 |
+
"eval_runtime": 33.8924,
|
1087 |
+
"eval_samples_per_second": 11.596,
|
1088 |
+
"eval_steps_per_second": 1.475,
|
1089 |
+
"step": 7200
|
1090 |
+
},
|
1091 |
+
{
|
1092 |
+
"epoch": 37.43589743589744,
|
1093 |
+
"grad_norm": 21.8422794342041,
|
1094 |
+
"learning_rate": 9.271000000000002e-06,
|
1095 |
+
"loss": 1.189,
|
1096 |
+
"step": 7300
|
1097 |
+
},
|
1098 |
+
{
|
1099 |
+
"epoch": 37.43589743589744,
|
1100 |
+
"eval_loss": 1.2060439586639404,
|
1101 |
+
"eval_runtime": 34.2775,
|
1102 |
+
"eval_samples_per_second": 11.465,
|
1103 |
+
"eval_steps_per_second": 1.459,
|
1104 |
+
"step": 7300
|
1105 |
+
},
|
1106 |
+
{
|
1107 |
+
"epoch": 37.94871794871795,
|
1108 |
+
"grad_norm": 28.315584182739258,
|
1109 |
+
"learning_rate": 9.261000000000002e-06,
|
1110 |
+
"loss": 1.139,
|
1111 |
+
"step": 7400
|
1112 |
+
},
|
1113 |
+
{
|
1114 |
+
"epoch": 37.94871794871795,
|
1115 |
+
"eval_loss": 1.191786527633667,
|
1116 |
+
"eval_runtime": 33.8013,
|
1117 |
+
"eval_samples_per_second": 11.627,
|
1118 |
+
"eval_steps_per_second": 1.479,
|
1119 |
+
"step": 7400
|
1120 |
+
},
|
1121 |
+
{
|
1122 |
+
"epoch": 38.46153846153846,
|
1123 |
+
"grad_norm": 30.92888641357422,
|
1124 |
+
"learning_rate": 9.2511e-06,
|
1125 |
+
"loss": 1.1509,
|
1126 |
+
"step": 7500
|
1127 |
+
},
|
1128 |
+
{
|
1129 |
+
"epoch": 38.46153846153846,
|
1130 |
+
"eval_loss": 1.1886085271835327,
|
1131 |
+
"eval_runtime": 33.8579,
|
1132 |
+
"eval_samples_per_second": 11.607,
|
1133 |
+
"eval_steps_per_second": 1.477,
|
1134 |
+
"step": 7500
|
1135 |
+
},
|
1136 |
+
{
|
1137 |
+
"epoch": 38.97435897435897,
|
1138 |
+
"grad_norm": 21.603288650512695,
|
1139 |
+
"learning_rate": 9.2411e-06,
|
1140 |
+
"loss": 1.1592,
|
1141 |
+
"step": 7600
|
1142 |
+
},
|
1143 |
+
{
|
1144 |
+
"epoch": 38.97435897435897,
|
1145 |
+
"eval_loss": 1.1912109851837158,
|
1146 |
+
"eval_runtime": 33.7856,
|
1147 |
+
"eval_samples_per_second": 11.632,
|
1148 |
+
"eval_steps_per_second": 1.48,
|
1149 |
+
"step": 7600
|
1150 |
+
},
|
1151 |
+
{
|
1152 |
+
"epoch": 39.48717948717949,
|
1153 |
+
"grad_norm": 42.09080123901367,
|
1154 |
+
"learning_rate": 9.2311e-06,
|
1155 |
+
"loss": 1.1501,
|
1156 |
+
"step": 7700
|
1157 |
+
},
|
1158 |
+
{
|
1159 |
+
"epoch": 39.48717948717949,
|
1160 |
+
"eval_loss": 1.1855125427246094,
|
1161 |
+
"eval_runtime": 33.7703,
|
1162 |
+
"eval_samples_per_second": 11.637,
|
1163 |
+
"eval_steps_per_second": 1.481,
|
1164 |
+
"step": 7700
|
1165 |
+
},
|
1166 |
+
{
|
1167 |
+
"epoch": 40.0,
|
1168 |
+
"grad_norm": 27.183298110961914,
|
1169 |
+
"learning_rate": 9.2211e-06,
|
1170 |
+
"loss": 1.1287,
|
1171 |
+
"step": 7800
|
1172 |
+
},
|
1173 |
+
{
|
1174 |
+
"epoch": 40.0,
|
1175 |
+
"eval_loss": 1.1904667615890503,
|
1176 |
+
"eval_runtime": 33.7348,
|
1177 |
+
"eval_samples_per_second": 11.65,
|
1178 |
+
"eval_steps_per_second": 1.482,
|
1179 |
+
"step": 7800
|
1180 |
+
},
|
1181 |
+
{
|
1182 |
+
"epoch": 40.51282051282051,
|
1183 |
+
"grad_norm": 62.908424377441406,
|
1184 |
+
"learning_rate": 9.2111e-06,
|
1185 |
+
"loss": 1.1598,
|
1186 |
+
"step": 7900
|
1187 |
+
},
|
1188 |
+
{
|
1189 |
+
"epoch": 40.51282051282051,
|
1190 |
+
"eval_loss": 1.1837190389633179,
|
1191 |
+
"eval_runtime": 34.0069,
|
1192 |
+
"eval_samples_per_second": 11.556,
|
1193 |
+
"eval_steps_per_second": 1.47,
|
1194 |
+
"step": 7900
|
1195 |
+
},
|
1196 |
+
{
|
1197 |
+
"epoch": 41.02564102564103,
|
1198 |
+
"grad_norm": 33.93372344970703,
|
1199 |
+
"learning_rate": 9.2011e-06,
|
1200 |
+
"loss": 1.1308,
|
1201 |
+
"step": 8000
|
1202 |
+
},
|
1203 |
+
{
|
1204 |
+
"epoch": 41.02564102564103,
|
1205 |
+
"eval_loss": 1.1925432682037354,
|
1206 |
+
"eval_runtime": 33.8881,
|
1207 |
+
"eval_samples_per_second": 11.597,
|
1208 |
+
"eval_steps_per_second": 1.475,
|
1209 |
+
"step": 8000
|
1210 |
+
},
|
1211 |
+
{
|
1212 |
+
"epoch": 41.53846153846154,
|
1213 |
+
"grad_norm": 45.57182312011719,
|
1214 |
+
"learning_rate": 9.1911e-06,
|
1215 |
+
"loss": 1.1496,
|
1216 |
+
"step": 8100
|
1217 |
+
},
|
1218 |
+
{
|
1219 |
+
"epoch": 41.53846153846154,
|
1220 |
+
"eval_loss": 1.204172968864441,
|
1221 |
+
"eval_runtime": 33.8945,
|
1222 |
+
"eval_samples_per_second": 11.595,
|
1223 |
+
"eval_steps_per_second": 1.475,
|
1224 |
+
"step": 8100
|
1225 |
+
},
|
1226 |
+
{
|
1227 |
+
"epoch": 42.05128205128205,
|
1228 |
+
"grad_norm": 25.79231834411621,
|
1229 |
+
"learning_rate": 9.181100000000001e-06,
|
1230 |
+
"loss": 1.098,
|
1231 |
+
"step": 8200
|
1232 |
+
},
|
1233 |
+
{
|
1234 |
+
"epoch": 42.05128205128205,
|
1235 |
+
"eval_loss": 1.1738271713256836,
|
1236 |
+
"eval_runtime": 34.3546,
|
1237 |
+
"eval_samples_per_second": 11.44,
|
1238 |
+
"eval_steps_per_second": 1.455,
|
1239 |
+
"step": 8200
|
1240 |
+
},
|
1241 |
+
{
|
1242 |
+
"epoch": 42.56410256410256,
|
1243 |
+
"grad_norm": 27.62971305847168,
|
1244 |
+
"learning_rate": 9.171100000000001e-06,
|
1245 |
+
"loss": 1.1275,
|
1246 |
+
"step": 8300
|
1247 |
+
},
|
1248 |
+
{
|
1249 |
+
"epoch": 42.56410256410256,
|
1250 |
+
"eval_loss": 1.1773978471755981,
|
1251 |
+
"eval_runtime": 33.7608,
|
1252 |
+
"eval_samples_per_second": 11.641,
|
1253 |
+
"eval_steps_per_second": 1.481,
|
1254 |
+
"step": 8300
|
1255 |
+
},
|
1256 |
+
{
|
1257 |
+
"epoch": 43.07692307692308,
|
1258 |
+
"grad_norm": 20.703998565673828,
|
1259 |
+
"learning_rate": 9.161100000000001e-06,
|
1260 |
+
"loss": 1.1524,
|
1261 |
+
"step": 8400
|
1262 |
+
},
|
1263 |
+
{
|
1264 |
+
"epoch": 43.07692307692308,
|
1265 |
+
"eval_loss": 1.1819504499435425,
|
1266 |
+
"eval_runtime": 33.9332,
|
1267 |
+
"eval_samples_per_second": 11.582,
|
1268 |
+
"eval_steps_per_second": 1.473,
|
1269 |
+
"step": 8400
|
1270 |
+
},
|
1271 |
+
{
|
1272 |
+
"epoch": 43.58974358974359,
|
1273 |
+
"grad_norm": 29.18483543395996,
|
1274 |
+
"learning_rate": 9.151100000000001e-06,
|
1275 |
+
"loss": 1.1022,
|
1276 |
+
"step": 8500
|
1277 |
+
},
|
1278 |
+
{
|
1279 |
+
"epoch": 43.58974358974359,
|
1280 |
+
"eval_loss": 1.1720554828643799,
|
1281 |
+
"eval_runtime": 33.872,
|
1282 |
+
"eval_samples_per_second": 11.602,
|
1283 |
+
"eval_steps_per_second": 1.476,
|
1284 |
+
"step": 8500
|
1285 |
+
},
|
1286 |
+
{
|
1287 |
+
"epoch": 44.1025641025641,
|
1288 |
+
"grad_norm": 34.14143371582031,
|
1289 |
+
"learning_rate": 9.141100000000001e-06,
|
1290 |
+
"loss": 1.1303,
|
1291 |
+
"step": 8600
|
1292 |
+
},
|
1293 |
+
{
|
1294 |
+
"epoch": 44.1025641025641,
|
1295 |
+
"eval_loss": 1.1805644035339355,
|
1296 |
+
"eval_runtime": 33.9223,
|
1297 |
+
"eval_samples_per_second": 11.585,
|
1298 |
+
"eval_steps_per_second": 1.474,
|
1299 |
+
"step": 8600
|
1300 |
+
},
|
1301 |
+
{
|
1302 |
+
"epoch": 44.61538461538461,
|
1303 |
+
"grad_norm": 43.17974853515625,
|
1304 |
+
"learning_rate": 9.1311e-06,
|
1305 |
+
"loss": 1.0919,
|
1306 |
+
"step": 8700
|
1307 |
+
},
|
1308 |
+
{
|
1309 |
+
"epoch": 44.61538461538461,
|
1310 |
+
"eval_loss": 1.1851396560668945,
|
1311 |
+
"eval_runtime": 33.9872,
|
1312 |
+
"eval_samples_per_second": 11.563,
|
1313 |
+
"eval_steps_per_second": 1.471,
|
1314 |
+
"step": 8700
|
1315 |
+
},
|
1316 |
+
{
|
1317 |
+
"epoch": 45.12820512820513,
|
1318 |
+
"grad_norm": 31.169639587402344,
|
1319 |
+
"learning_rate": 9.1211e-06,
|
1320 |
+
"loss": 1.1156,
|
1321 |
+
"step": 8800
|
1322 |
+
},
|
1323 |
+
{
|
1324 |
+
"epoch": 45.12820512820513,
|
1325 |
+
"eval_loss": 1.1849329471588135,
|
1326 |
+
"eval_runtime": 33.8263,
|
1327 |
+
"eval_samples_per_second": 11.618,
|
1328 |
+
"eval_steps_per_second": 1.478,
|
1329 |
+
"step": 8800
|
1330 |
+
},
|
1331 |
+
{
|
1332 |
+
"epoch": 45.64102564102564,
|
1333 |
+
"grad_norm": 47.28102493286133,
|
1334 |
+
"learning_rate": 9.1111e-06,
|
1335 |
+
"loss": 1.1201,
|
1336 |
+
"step": 8900
|
1337 |
+
},
|
1338 |
+
{
|
1339 |
+
"epoch": 45.64102564102564,
|
1340 |
+
"eval_loss": 1.1575498580932617,
|
1341 |
+
"eval_runtime": 33.7801,
|
1342 |
+
"eval_samples_per_second": 11.634,
|
1343 |
+
"eval_steps_per_second": 1.48,
|
1344 |
+
"step": 8900
|
1345 |
+
},
|
1346 |
+
{
|
1347 |
+
"epoch": 46.15384615384615,
|
1348 |
+
"grad_norm": 62.14609146118164,
|
1349 |
+
"learning_rate": 9.1011e-06,
|
1350 |
+
"loss": 1.0987,
|
1351 |
+
"step": 9000
|
1352 |
+
},
|
1353 |
+
{
|
1354 |
+
"epoch": 46.15384615384615,
|
1355 |
+
"eval_loss": 1.16815185546875,
|
1356 |
+
"eval_runtime": 33.7678,
|
1357 |
+
"eval_samples_per_second": 11.638,
|
1358 |
+
"eval_steps_per_second": 1.481,
|
1359 |
+
"step": 9000
|
1360 |
+
},
|
1361 |
+
{
|
1362 |
+
"epoch": 46.666666666666664,
|
1363 |
+
"grad_norm": 77.31022644042969,
|
1364 |
+
"learning_rate": 9.0911e-06,
|
1365 |
+
"loss": 1.119,
|
1366 |
+
"step": 9100
|
1367 |
+
},
|
1368 |
+
{
|
1369 |
+
"epoch": 46.666666666666664,
|
1370 |
+
"eval_loss": 1.1787101030349731,
|
1371 |
+
"eval_runtime": 34.2194,
|
1372 |
+
"eval_samples_per_second": 11.485,
|
1373 |
+
"eval_steps_per_second": 1.461,
|
1374 |
+
"step": 9100
|
1375 |
+
},
|
1376 |
+
{
|
1377 |
+
"epoch": 47.17948717948718,
|
1378 |
+
"grad_norm": 17.737777709960938,
|
1379 |
+
"learning_rate": 9.0811e-06,
|
1380 |
+
"loss": 1.1152,
|
1381 |
+
"step": 9200
|
1382 |
+
},
|
1383 |
+
{
|
1384 |
+
"epoch": 47.17948717948718,
|
1385 |
+
"eval_loss": 1.1642876863479614,
|
1386 |
+
"eval_runtime": 33.9617,
|
1387 |
+
"eval_samples_per_second": 11.572,
|
1388 |
+
"eval_steps_per_second": 1.472,
|
1389 |
+
"step": 9200
|
1390 |
+
},
|
1391 |
+
{
|
1392 |
+
"epoch": 47.69230769230769,
|
1393 |
+
"grad_norm": 22.58513069152832,
|
1394 |
+
"learning_rate": 9.0711e-06,
|
1395 |
+
"loss": 1.0801,
|
1396 |
+
"step": 9300
|
1397 |
+
},
|
1398 |
+
{
|
1399 |
+
"epoch": 47.69230769230769,
|
1400 |
+
"eval_loss": 1.1710031032562256,
|
1401 |
+
"eval_runtime": 33.6338,
|
1402 |
+
"eval_samples_per_second": 11.685,
|
1403 |
+
"eval_steps_per_second": 1.487,
|
1404 |
+
"step": 9300
|
1405 |
+
},
|
1406 |
+
{
|
1407 |
+
"epoch": 48.205128205128204,
|
1408 |
+
"grad_norm": 17.689064025878906,
|
1409 |
+
"learning_rate": 9.0611e-06,
|
1410 |
+
"loss": 1.1669,
|
1411 |
+
"step": 9400
|
1412 |
+
},
|
1413 |
+
{
|
1414 |
+
"epoch": 48.205128205128204,
|
1415 |
+
"eval_loss": 1.1800153255462646,
|
1416 |
+
"eval_runtime": 33.5397,
|
1417 |
+
"eval_samples_per_second": 11.717,
|
1418 |
+
"eval_steps_per_second": 1.491,
|
1419 |
+
"step": 9400
|
1420 |
+
},
|
1421 |
+
{
|
1422 |
+
"epoch": 48.717948717948715,
|
1423 |
+
"grad_norm": 20.76385498046875,
|
1424 |
+
"learning_rate": 9.0511e-06,
|
1425 |
+
"loss": 1.0969,
|
1426 |
+
"step": 9500
|
1427 |
+
},
|
1428 |
+
{
|
1429 |
+
"epoch": 48.717948717948715,
|
1430 |
+
"eval_loss": 1.170789361000061,
|
1431 |
+
"eval_runtime": 33.4901,
|
1432 |
+
"eval_samples_per_second": 11.735,
|
1433 |
+
"eval_steps_per_second": 1.493,
|
1434 |
+
"step": 9500
|
1435 |
+
},
|
1436 |
+
{
|
1437 |
+
"epoch": 49.23076923076923,
|
1438 |
+
"grad_norm": 53.716548919677734,
|
1439 |
+
"learning_rate": 9.0411e-06,
|
1440 |
+
"loss": 1.1053,
|
1441 |
+
"step": 9600
|
1442 |
+
},
|
1443 |
+
{
|
1444 |
+
"epoch": 49.23076923076923,
|
1445 |
+
"eval_loss": 1.1888952255249023,
|
1446 |
+
"eval_runtime": 33.3262,
|
1447 |
+
"eval_samples_per_second": 11.793,
|
1448 |
+
"eval_steps_per_second": 1.5,
|
1449 |
+
"step": 9600
|
1450 |
+
},
|
1451 |
+
{
|
1452 |
+
"epoch": 49.743589743589745,
|
1453 |
+
"grad_norm": 70.57758331298828,
|
1454 |
+
"learning_rate": 9.0311e-06,
|
1455 |
+
"loss": 1.088,
|
1456 |
+
"step": 9700
|
1457 |
+
},
|
1458 |
+
{
|
1459 |
+
"epoch": 49.743589743589745,
|
1460 |
+
"eval_loss": 1.1654279232025146,
|
1461 |
+
"eval_runtime": 33.4095,
|
1462 |
+
"eval_samples_per_second": 11.763,
|
1463 |
+
"eval_steps_per_second": 1.497,
|
1464 |
+
"step": 9700
|
1465 |
+
},
|
1466 |
+
{
|
1467 |
+
"epoch": 50.256410256410255,
|
1468 |
+
"grad_norm": 48.835018157958984,
|
1469 |
+
"learning_rate": 9.0211e-06,
|
1470 |
+
"loss": 1.1185,
|
1471 |
+
"step": 9800
|
1472 |
+
},
|
1473 |
+
{
|
1474 |
+
"epoch": 50.256410256410255,
|
1475 |
+
"eval_loss": 1.174896240234375,
|
1476 |
+
"eval_runtime": 33.655,
|
1477 |
+
"eval_samples_per_second": 11.677,
|
1478 |
+
"eval_steps_per_second": 1.486,
|
1479 |
+
"step": 9800
|
1480 |
+
},
|
1481 |
+
{
|
1482 |
+
"epoch": 50.76923076923077,
|
1483 |
+
"grad_norm": 27.838232040405273,
|
1484 |
+
"learning_rate": 9.011100000000001e-06,
|
1485 |
+
"loss": 1.0741,
|
1486 |
+
"step": 9900
|
1487 |
+
},
|
1488 |
+
{
|
1489 |
+
"epoch": 50.76923076923077,
|
1490 |
+
"eval_loss": 1.1926134824752808,
|
1491 |
+
"eval_runtime": 33.5437,
|
1492 |
+
"eval_samples_per_second": 11.716,
|
1493 |
+
"eval_steps_per_second": 1.491,
|
1494 |
+
"step": 9900
|
1495 |
+
},
|
1496 |
+
{
|
1497 |
+
"epoch": 51.282051282051285,
|
1498 |
+
"grad_norm": 36.455204010009766,
|
1499 |
+
"learning_rate": 9.001100000000001e-06,
|
1500 |
+
"loss": 1.1212,
|
1501 |
+
"step": 10000
|
1502 |
+
},
|
1503 |
+
{
|
1504 |
+
"epoch": 51.282051282051285,
|
1505 |
+
"eval_loss": 1.1525243520736694,
|
1506 |
+
"eval_runtime": 33.5794,
|
1507 |
+
"eval_samples_per_second": 11.704,
|
1508 |
+
"eval_steps_per_second": 1.489,
|
1509 |
+
"step": 10000
|
1510 |
+
},
|
1511 |
+
{
|
1512 |
+
"epoch": 51.794871794871796,
|
1513 |
+
"grad_norm": 24.63564109802246,
|
1514 |
+
"learning_rate": 8.991100000000001e-06,
|
1515 |
+
"loss": 1.0958,
|
1516 |
+
"step": 10100
|
1517 |
+
},
|
1518 |
+
{
|
1519 |
+
"epoch": 51.794871794871796,
|
1520 |
+
"eval_loss": 1.14644193649292,
|
1521 |
+
"eval_runtime": 33.4871,
|
1522 |
+
"eval_samples_per_second": 11.736,
|
1523 |
+
"eval_steps_per_second": 1.493,
|
1524 |
+
"step": 10100
|
1525 |
+
},
|
1526 |
+
{
|
1527 |
+
"epoch": 52.30769230769231,
|
1528 |
+
"grad_norm": 52.96881103515625,
|
1529 |
+
"learning_rate": 8.981100000000001e-06,
|
1530 |
+
"loss": 1.0793,
|
1531 |
+
"step": 10200
|
1532 |
+
},
|
1533 |
+
{
|
1534 |
+
"epoch": 52.30769230769231,
|
1535 |
+
"eval_loss": 1.154321312904358,
|
1536 |
+
"eval_runtime": 33.4968,
|
1537 |
+
"eval_samples_per_second": 11.732,
|
1538 |
+
"eval_steps_per_second": 1.493,
|
1539 |
+
"step": 10200
|
1540 |
+
},
|
1541 |
+
{
|
1542 |
+
"epoch": 52.82051282051282,
|
1543 |
+
"grad_norm": 24.98472023010254,
|
1544 |
+
"learning_rate": 8.9711e-06,
|
1545 |
+
"loss": 1.1111,
|
1546 |
+
"step": 10300
|
1547 |
+
},
|
1548 |
+
{
|
1549 |
+
"epoch": 52.82051282051282,
|
1550 |
+
"eval_loss": 1.1450814008712769,
|
1551 |
+
"eval_runtime": 33.7441,
|
1552 |
+
"eval_samples_per_second": 11.646,
|
1553 |
+
"eval_steps_per_second": 1.482,
|
1554 |
+
"step": 10300
|
1555 |
+
},
|
1556 |
+
{
|
1557 |
+
"epoch": 53.333333333333336,
|
1558 |
+
"grad_norm": 37.849769592285156,
|
1559 |
+
"learning_rate": 8.9611e-06,
|
1560 |
+
"loss": 1.0579,
|
1561 |
+
"step": 10400
|
1562 |
+
},
|
1563 |
+
{
|
1564 |
+
"epoch": 53.333333333333336,
|
1565 |
+
"eval_loss": 1.1465113162994385,
|
1566 |
+
"eval_runtime": 33.6742,
|
1567 |
+
"eval_samples_per_second": 11.671,
|
1568 |
+
"eval_steps_per_second": 1.485,
|
1569 |
+
"step": 10400
|
1570 |
+
},
|
1571 |
+
{
|
1572 |
+
"epoch": 53.84615384615385,
|
1573 |
+
"grad_norm": 32.475791931152344,
|
1574 |
+
"learning_rate": 8.9511e-06,
|
1575 |
+
"loss": 1.0959,
|
1576 |
+
"step": 10500
|
1577 |
+
},
|
1578 |
+
{
|
1579 |
+
"epoch": 53.84615384615385,
|
1580 |
+
"eval_loss": 1.1625763177871704,
|
1581 |
+
"eval_runtime": 33.5922,
|
1582 |
+
"eval_samples_per_second": 11.699,
|
1583 |
+
"eval_steps_per_second": 1.488,
|
1584 |
+
"step": 10500
|
1585 |
+
},
|
1586 |
+
{
|
1587 |
+
"epoch": 54.35897435897436,
|
1588 |
+
"grad_norm": 29.02692413330078,
|
1589 |
+
"learning_rate": 8.9411e-06,
|
1590 |
+
"loss": 1.1046,
|
1591 |
+
"step": 10600
|
1592 |
+
},
|
1593 |
+
{
|
1594 |
+
"epoch": 54.35897435897436,
|
1595 |
+
"eval_loss": 1.161304235458374,
|
1596 |
+
"eval_runtime": 33.4954,
|
1597 |
+
"eval_samples_per_second": 11.733,
|
1598 |
+
"eval_steps_per_second": 1.493,
|
1599 |
+
"step": 10600
|
1600 |
+
},
|
1601 |
+
{
|
1602 |
+
"epoch": 54.87179487179487,
|
1603 |
+
"grad_norm": 93.73539733886719,
|
1604 |
+
"learning_rate": 8.9311e-06,
|
1605 |
+
"loss": 1.1034,
|
1606 |
+
"step": 10700
|
1607 |
+
},
|
1608 |
+
{
|
1609 |
+
"epoch": 54.87179487179487,
|
1610 |
+
"eval_loss": 1.175752878189087,
|
1611 |
+
"eval_runtime": 33.5496,
|
1612 |
+
"eval_samples_per_second": 11.714,
|
1613 |
+
"eval_steps_per_second": 1.49,
|
1614 |
+
"step": 10700
|
1615 |
+
},
|
1616 |
+
{
|
1617 |
+
"epoch": 55.38461538461539,
|
1618 |
+
"grad_norm": 23.342376708984375,
|
1619 |
+
"learning_rate": 8.9211e-06,
|
1620 |
+
"loss": 1.0699,
|
1621 |
+
"step": 10800
|
1622 |
+
},
|
1623 |
+
{
|
1624 |
+
"epoch": 55.38461538461539,
|
1625 |
+
"eval_loss": 1.1697540283203125,
|
1626 |
+
"eval_runtime": 33.468,
|
1627 |
+
"eval_samples_per_second": 11.743,
|
1628 |
+
"eval_steps_per_second": 1.494,
|
1629 |
+
"step": 10800
|
1630 |
+
},
|
1631 |
+
{
|
1632 |
+
"epoch": 55.8974358974359,
|
1633 |
+
"grad_norm": 55.842430114746094,
|
1634 |
+
"learning_rate": 8.9111e-06,
|
1635 |
+
"loss": 1.0773,
|
1636 |
+
"step": 10900
|
1637 |
+
},
|
1638 |
+
{
|
1639 |
+
"epoch": 55.8974358974359,
|
1640 |
+
"eval_loss": 1.151752233505249,
|
1641 |
+
"eval_runtime": 33.9176,
|
1642 |
+
"eval_samples_per_second": 11.587,
|
1643 |
+
"eval_steps_per_second": 1.474,
|
1644 |
+
"step": 10900
|
1645 |
+
},
|
1646 |
+
{
|
1647 |
+
"epoch": 56.41025641025641,
|
1648 |
+
"grad_norm": 34.547996520996094,
|
1649 |
+
"learning_rate": 8.9012e-06,
|
1650 |
+
"loss": 1.0888,
|
1651 |
+
"step": 11000
|
1652 |
+
},
|
1653 |
+
{
|
1654 |
+
"epoch": 56.41025641025641,
|
1655 |
+
"eval_loss": 1.1507455110549927,
|
1656 |
+
"eval_runtime": 33.5018,
|
1657 |
+
"eval_samples_per_second": 11.731,
|
1658 |
+
"eval_steps_per_second": 1.492,
|
1659 |
+
"step": 11000
|
1660 |
+
},
|
1661 |
+
{
|
1662 |
+
"epoch": 56.92307692307692,
|
1663 |
+
"grad_norm": 24.540210723876953,
|
1664 |
+
"learning_rate": 8.8912e-06,
|
1665 |
+
"loss": 1.0634,
|
1666 |
+
"step": 11100
|
1667 |
+
},
|
1668 |
+
{
|
1669 |
+
"epoch": 56.92307692307692,
|
1670 |
+
"eval_loss": 1.1604851484298706,
|
1671 |
+
"eval_runtime": 33.5283,
|
1672 |
+
"eval_samples_per_second": 11.721,
|
1673 |
+
"eval_steps_per_second": 1.491,
|
1674 |
+
"step": 11100
|
1675 |
+
},
|
1676 |
+
{
|
1677 |
+
"epoch": 57.43589743589744,
|
1678 |
+
"grad_norm": 25.421348571777344,
|
1679 |
+
"learning_rate": 8.8812e-06,
|
1680 |
+
"loss": 1.1177,
|
1681 |
+
"step": 11200
|
1682 |
+
},
|
1683 |
+
{
|
1684 |
+
"epoch": 57.43589743589744,
|
1685 |
+
"eval_loss": 1.1580452919006348,
|
1686 |
+
"eval_runtime": 33.4736,
|
1687 |
+
"eval_samples_per_second": 11.741,
|
1688 |
+
"eval_steps_per_second": 1.494,
|
1689 |
+
"step": 11200
|
1690 |
+
},
|
1691 |
+
{
|
1692 |
+
"epoch": 57.94871794871795,
|
1693 |
+
"grad_norm": 74.44795989990234,
|
1694 |
+
"learning_rate": 8.8712e-06,
|
1695 |
+
"loss": 1.0649,
|
1696 |
+
"step": 11300
|
1697 |
+
},
|
1698 |
+
{
|
1699 |
+
"epoch": 57.94871794871795,
|
1700 |
+
"eval_loss": 1.165457010269165,
|
1701 |
+
"eval_runtime": 33.477,
|
1702 |
+
"eval_samples_per_second": 11.739,
|
1703 |
+
"eval_steps_per_second": 1.494,
|
1704 |
+
"step": 11300
|
1705 |
+
},
|
1706 |
+
{
|
1707 |
+
"epoch": 58.46153846153846,
|
1708 |
+
"grad_norm": 21.34556007385254,
|
1709 |
+
"learning_rate": 8.8612e-06,
|
1710 |
+
"loss": 1.0379,
|
1711 |
+
"step": 11400
|
1712 |
+
},
|
1713 |
+
{
|
1714 |
+
"epoch": 58.46153846153846,
|
1715 |
+
"eval_loss": 1.156040072441101,
|
1716 |
+
"eval_runtime": 33.5737,
|
1717 |
+
"eval_samples_per_second": 11.706,
|
1718 |
+
"eval_steps_per_second": 1.489,
|
1719 |
+
"step": 11400
|
1720 |
+
},
|
1721 |
+
{
|
1722 |
+
"epoch": 58.97435897435897,
|
1723 |
+
"grad_norm": 27.036218643188477,
|
1724 |
+
"learning_rate": 8.851200000000001e-06,
|
1725 |
+
"loss": 1.089,
|
1726 |
+
"step": 11500
|
1727 |
+
},
|
1728 |
+
{
|
1729 |
+
"epoch": 58.97435897435897,
|
1730 |
+
"eval_loss": 1.134421944618225,
|
1731 |
+
"eval_runtime": 33.5208,
|
1732 |
+
"eval_samples_per_second": 11.724,
|
1733 |
+
"eval_steps_per_second": 1.492,
|
1734 |
+
"step": 11500
|
1735 |
+
},
|
1736 |
+
{
|
1737 |
+
"epoch": 59.48717948717949,
|
1738 |
+
"grad_norm": 18.133159637451172,
|
1739 |
+
"learning_rate": 8.841200000000001e-06,
|
1740 |
+
"loss": 1.1044,
|
1741 |
+
"step": 11600
|
1742 |
+
},
|
1743 |
+
{
|
1744 |
+
"epoch": 59.48717948717949,
|
1745 |
+
"eval_loss": 1.1284948587417603,
|
1746 |
+
"eval_runtime": 33.4738,
|
1747 |
+
"eval_samples_per_second": 11.741,
|
1748 |
+
"eval_steps_per_second": 1.494,
|
1749 |
+
"step": 11600
|
1750 |
+
},
|
1751 |
+
{
|
1752 |
+
"epoch": 60.0,
|
1753 |
+
"grad_norm": 51.20466613769531,
|
1754 |
+
"learning_rate": 8.831200000000001e-06,
|
1755 |
+
"loss": 1.021,
|
1756 |
+
"step": 11700
|
1757 |
+
},
|
1758 |
+
{
|
1759 |
+
"epoch": 60.0,
|
1760 |
+
"eval_loss": 1.1374192237854004,
|
1761 |
+
"eval_runtime": 33.3528,
|
1762 |
+
"eval_samples_per_second": 11.783,
|
1763 |
+
"eval_steps_per_second": 1.499,
|
1764 |
+
"step": 11700
|
1765 |
+
},
|
1766 |
+
{
|
1767 |
+
"epoch": 60.51282051282051,
|
1768 |
+
"grad_norm": 20.16541290283203,
|
1769 |
+
"learning_rate": 8.821200000000001e-06,
|
1770 |
+
"loss": 1.0547,
|
1771 |
+
"step": 11800
|
1772 |
+
},
|
1773 |
+
{
|
1774 |
+
"epoch": 60.51282051282051,
|
1775 |
+
"eval_loss": 1.1408305168151855,
|
1776 |
+
"eval_runtime": 33.9564,
|
1777 |
+
"eval_samples_per_second": 11.574,
|
1778 |
+
"eval_steps_per_second": 1.472,
|
1779 |
+
"step": 11800
|
1780 |
+
},
|
1781 |
+
{
|
1782 |
+
"epoch": 61.02564102564103,
|
1783 |
+
"grad_norm": 22.255817413330078,
|
1784 |
+
"learning_rate": 8.811200000000001e-06,
|
1785 |
+
"loss": 1.0737,
|
1786 |
+
"step": 11900
|
1787 |
+
},
|
1788 |
+
{
|
1789 |
+
"epoch": 61.02564102564103,
|
1790 |
+
"eval_loss": 1.1481328010559082,
|
1791 |
+
"eval_runtime": 33.4272,
|
1792 |
+
"eval_samples_per_second": 11.757,
|
1793 |
+
"eval_steps_per_second": 1.496,
|
1794 |
+
"step": 11900
|
1795 |
+
},
|
1796 |
+
{
|
1797 |
+
"epoch": 61.53846153846154,
|
1798 |
+
"grad_norm": 34.67921447753906,
|
1799 |
+
"learning_rate": 8.801200000000001e-06,
|
1800 |
+
"loss": 1.0152,
|
1801 |
+
"step": 12000
|
1802 |
+
},
|
1803 |
+
{
|
1804 |
+
"epoch": 61.53846153846154,
|
1805 |
+
"eval_loss": 1.1518502235412598,
|
1806 |
+
"eval_runtime": 33.4288,
|
1807 |
+
"eval_samples_per_second": 11.756,
|
1808 |
+
"eval_steps_per_second": 1.496,
|
1809 |
+
"step": 12000
|
1810 |
+
},
|
1811 |
+
{
|
1812 |
+
"epoch": 62.05128205128205,
|
1813 |
+
"grad_norm": 52.76307678222656,
|
1814 |
+
"learning_rate": 8.791200000000001e-06,
|
1815 |
+
"loss": 1.0853,
|
1816 |
+
"step": 12100
|
1817 |
+
},
|
1818 |
+
{
|
1819 |
+
"epoch": 62.05128205128205,
|
1820 |
+
"eval_loss": 1.1410694122314453,
|
1821 |
+
"eval_runtime": 33.2178,
|
1822 |
+
"eval_samples_per_second": 11.831,
|
1823 |
+
"eval_steps_per_second": 1.505,
|
1824 |
+
"step": 12100
|
1825 |
+
},
|
1826 |
+
{
|
1827 |
+
"epoch": 62.56410256410256,
|
1828 |
+
"grad_norm": 36.955753326416016,
|
1829 |
+
"learning_rate": 8.781200000000002e-06,
|
1830 |
+
"loss": 1.0754,
|
1831 |
+
"step": 12200
|
1832 |
+
},
|
1833 |
+
{
|
1834 |
+
"epoch": 62.56410256410256,
|
1835 |
+
"eval_loss": 1.1417096853256226,
|
1836 |
+
"eval_runtime": 33.6371,
|
1837 |
+
"eval_samples_per_second": 11.684,
|
1838 |
+
"eval_steps_per_second": 1.486,
|
1839 |
+
"step": 12200
|
1840 |
+
},
|
1841 |
+
{
|
1842 |
+
"epoch": 63.07692307692308,
|
1843 |
+
"grad_norm": 30.68532371520996,
|
1844 |
+
"learning_rate": 8.7712e-06,
|
1845 |
+
"loss": 1.0177,
|
1846 |
+
"step": 12300
|
1847 |
+
},
|
1848 |
+
{
|
1849 |
+
"epoch": 63.07692307692308,
|
1850 |
+
"eval_loss": 1.1374719142913818,
|
1851 |
+
"eval_runtime": 33.4002,
|
1852 |
+
"eval_samples_per_second": 11.766,
|
1853 |
+
"eval_steps_per_second": 1.497,
|
1854 |
+
"step": 12300
|
1855 |
+
},
|
1856 |
+
{
|
1857 |
+
"epoch": 63.58974358974359,
|
1858 |
+
"grad_norm": 73.53378295898438,
|
1859 |
+
"learning_rate": 8.7612e-06,
|
1860 |
+
"loss": 1.0533,
|
1861 |
+
"step": 12400
|
1862 |
+
},
|
1863 |
+
{
|
1864 |
+
"epoch": 63.58974358974359,
|
1865 |
+
"eval_loss": 1.1490620374679565,
|
1866 |
+
"eval_runtime": 33.4237,
|
1867 |
+
"eval_samples_per_second": 11.758,
|
1868 |
+
"eval_steps_per_second": 1.496,
|
1869 |
+
"step": 12400
|
1870 |
+
},
|
1871 |
+
{
|
1872 |
+
"epoch": 64.1025641025641,
|
1873 |
+
"grad_norm": 29.064855575561523,
|
1874 |
+
"learning_rate": 8.7512e-06,
|
1875 |
+
"loss": 1.0359,
|
1876 |
+
"step": 12500
|
1877 |
+
},
|
1878 |
+
{
|
1879 |
+
"epoch": 64.1025641025641,
|
1880 |
+
"eval_loss": 1.1465009450912476,
|
1881 |
+
"eval_runtime": 33.4463,
|
1882 |
+
"eval_samples_per_second": 11.75,
|
1883 |
+
"eval_steps_per_second": 1.495,
|
1884 |
+
"step": 12500
|
1885 |
+
},
|
1886 |
+
{
|
1887 |
+
"epoch": 64.61538461538461,
|
1888 |
+
"grad_norm": 43.581565856933594,
|
1889 |
+
"learning_rate": 8.7412e-06,
|
1890 |
+
"loss": 1.0222,
|
1891 |
+
"step": 12600
|
1892 |
+
},
|
1893 |
+
{
|
1894 |
+
"epoch": 64.61538461538461,
|
1895 |
+
"eval_loss": 1.1300017833709717,
|
1896 |
+
"eval_runtime": 33.397,
|
1897 |
+
"eval_samples_per_second": 11.768,
|
1898 |
+
"eval_steps_per_second": 1.497,
|
1899 |
+
"step": 12600
|
1900 |
+
},
|
1901 |
+
{
|
1902 |
+
"epoch": 65.12820512820512,
|
1903 |
+
"grad_norm": 37.94940185546875,
|
1904 |
+
"learning_rate": 8.7312e-06,
|
1905 |
+
"loss": 1.0676,
|
1906 |
+
"step": 12700
|
1907 |
+
},
|
1908 |
+
{
|
1909 |
+
"epoch": 65.12820512820512,
|
1910 |
+
"eval_loss": 1.1514735221862793,
|
1911 |
+
"eval_runtime": 33.4854,
|
1912 |
+
"eval_samples_per_second": 11.736,
|
1913 |
+
"eval_steps_per_second": 1.493,
|
1914 |
+
"step": 12700
|
1915 |
+
},
|
1916 |
+
{
|
1917 |
+
"epoch": 65.64102564102564,
|
1918 |
+
"grad_norm": 19.844234466552734,
|
1919 |
+
"learning_rate": 8.7212e-06,
|
1920 |
+
"loss": 1.0438,
|
1921 |
+
"step": 12800
|
1922 |
+
},
|
1923 |
+
{
|
1924 |
+
"epoch": 65.64102564102564,
|
1925 |
+
"eval_loss": 1.166193962097168,
|
1926 |
+
"eval_runtime": 33.342,
|
1927 |
+
"eval_samples_per_second": 11.787,
|
1928 |
+
"eval_steps_per_second": 1.5,
|
1929 |
+
"step": 12800
|
1930 |
+
},
|
1931 |
+
{
|
1932 |
+
"epoch": 66.15384615384616,
|
1933 |
+
"grad_norm": 34.03626251220703,
|
1934 |
+
"learning_rate": 8.7112e-06,
|
1935 |
+
"loss": 0.9963,
|
1936 |
+
"step": 12900
|
1937 |
+
},
|
1938 |
+
{
|
1939 |
+
"epoch": 66.15384615384616,
|
1940 |
+
"eval_loss": 1.14573335647583,
|
1941 |
+
"eval_runtime": 33.5502,
|
1942 |
+
"eval_samples_per_second": 11.714,
|
1943 |
+
"eval_steps_per_second": 1.49,
|
1944 |
+
"step": 12900
|
1945 |
+
},
|
1946 |
+
{
|
1947 |
+
"epoch": 66.66666666666667,
|
1948 |
+
"grad_norm": 42.71162033081055,
|
1949 |
+
"learning_rate": 8.7012e-06,
|
1950 |
+
"loss": 1.0168,
|
1951 |
+
"step": 13000
|
1952 |
+
},
|
1953 |
+
{
|
1954 |
+
"epoch": 66.66666666666667,
|
1955 |
+
"eval_loss": 1.1428903341293335,
|
1956 |
+
"eval_runtime": 33.6161,
|
1957 |
+
"eval_samples_per_second": 11.691,
|
1958 |
+
"eval_steps_per_second": 1.487,
|
1959 |
+
"step": 13000
|
1960 |
+
},
|
1961 |
+
{
|
1962 |
+
"epoch": 67.17948717948718,
|
1963 |
+
"grad_norm": 44.0084114074707,
|
1964 |
+
"learning_rate": 8.6912e-06,
|
1965 |
+
"loss": 1.0713,
|
1966 |
+
"step": 13100
|
1967 |
+
},
|
1968 |
+
{
|
1969 |
+
"epoch": 67.17948717948718,
|
1970 |
+
"eval_loss": 1.1410739421844482,
|
1971 |
+
"eval_runtime": 33.5446,
|
1972 |
+
"eval_samples_per_second": 11.716,
|
1973 |
+
"eval_steps_per_second": 1.491,
|
1974 |
+
"step": 13100
|
1975 |
+
},
|
1976 |
+
{
|
1977 |
+
"epoch": 67.6923076923077,
|
1978 |
+
"grad_norm": 24.194211959838867,
|
1979 |
+
"learning_rate": 8.6812e-06,
|
1980 |
+
"loss": 1.0132,
|
1981 |
+
"step": 13200
|
1982 |
+
},
|
1983 |
+
{
|
1984 |
+
"epoch": 67.6923076923077,
|
1985 |
+
"eval_loss": 1.1532968282699585,
|
1986 |
+
"eval_runtime": 33.4966,
|
1987 |
+
"eval_samples_per_second": 11.733,
|
1988 |
+
"eval_steps_per_second": 1.493,
|
1989 |
+
"step": 13200
|
1990 |
+
},
|
1991 |
+
{
|
1992 |
+
"epoch": 68.2051282051282,
|
1993 |
+
"grad_norm": 14.505328178405762,
|
1994 |
+
"learning_rate": 8.671200000000001e-06,
|
1995 |
+
"loss": 1.0685,
|
1996 |
+
"step": 13300
|
1997 |
+
},
|
1998 |
+
{
|
1999 |
+
"epoch": 68.2051282051282,
|
2000 |
+
"eval_loss": 1.1466225385665894,
|
2001 |
+
"eval_runtime": 33.4196,
|
2002 |
+
"eval_samples_per_second": 11.76,
|
2003 |
+
"eval_steps_per_second": 1.496,
|
2004 |
+
"step": 13300
|
2005 |
+
},
|
2006 |
+
{
|
2007 |
+
"epoch": 68.71794871794872,
|
2008 |
+
"grad_norm": 35.94062423706055,
|
2009 |
+
"learning_rate": 8.661200000000001e-06,
|
2010 |
+
"loss": 1.0035,
|
2011 |
+
"step": 13400
|
2012 |
+
},
|
2013 |
+
{
|
2014 |
+
"epoch": 68.71794871794872,
|
2015 |
+
"eval_loss": 1.1296441555023193,
|
2016 |
+
"eval_runtime": 33.6415,
|
2017 |
+
"eval_samples_per_second": 11.682,
|
2018 |
+
"eval_steps_per_second": 1.486,
|
2019 |
+
"step": 13400
|
2020 |
+
},
|
2021 |
+
{
|
2022 |
+
"epoch": 69.23076923076923,
|
2023 |
+
"grad_norm": 24.583995819091797,
|
2024 |
+
"learning_rate": 8.651200000000001e-06,
|
2025 |
+
"loss": 1.0128,
|
2026 |
+
"step": 13500
|
2027 |
+
},
|
2028 |
+
{
|
2029 |
+
"epoch": 69.23076923076923,
|
2030 |
+
"eval_loss": 1.107634425163269,
|
2031 |
+
"eval_runtime": 33.7082,
|
2032 |
+
"eval_samples_per_second": 11.659,
|
2033 |
+
"eval_steps_per_second": 1.483,
|
2034 |
+
"step": 13500
|
2035 |
+
},
|
2036 |
+
{
|
2037 |
+
"epoch": 69.74358974358974,
|
2038 |
+
"grad_norm": 31.751113891601562,
|
2039 |
+
"learning_rate": 8.6413e-06,
|
2040 |
+
"loss": 1.0426,
|
2041 |
+
"step": 13600
|
2042 |
+
},
|
2043 |
+
{
|
2044 |
+
"epoch": 69.74358974358974,
|
2045 |
+
"eval_loss": 1.119456171989441,
|
2046 |
+
"eval_runtime": 33.5332,
|
2047 |
+
"eval_samples_per_second": 11.72,
|
2048 |
+
"eval_steps_per_second": 1.491,
|
2049 |
+
"step": 13600
|
2050 |
+
},
|
2051 |
+
{
|
2052 |
+
"epoch": 70.25641025641026,
|
2053 |
+
"grad_norm": 76.94485473632812,
|
2054 |
+
"learning_rate": 8.6313e-06,
|
2055 |
+
"loss": 1.0466,
|
2056 |
+
"step": 13700
|
2057 |
+
},
|
2058 |
+
{
|
2059 |
+
"epoch": 70.25641025641026,
|
2060 |
+
"eval_loss": 1.1407954692840576,
|
2061 |
+
"eval_runtime": 33.3405,
|
2062 |
+
"eval_samples_per_second": 11.787,
|
2063 |
+
"eval_steps_per_second": 1.5,
|
2064 |
+
"step": 13700
|
2065 |
+
},
|
2066 |
+
{
|
2067 |
+
"epoch": 70.76923076923077,
|
2068 |
+
"grad_norm": 36.27717590332031,
|
2069 |
+
"learning_rate": 8.6213e-06,
|
2070 |
+
"loss": 1.0371,
|
2071 |
+
"step": 13800
|
2072 |
+
},
|
2073 |
+
{
|
2074 |
+
"epoch": 70.76923076923077,
|
2075 |
+
"eval_loss": 1.1318871974945068,
|
2076 |
+
"eval_runtime": 33.4695,
|
2077 |
+
"eval_samples_per_second": 11.742,
|
2078 |
+
"eval_steps_per_second": 1.494,
|
2079 |
+
"step": 13800
|
2080 |
+
},
|
2081 |
+
{
|
2082 |
+
"epoch": 71.28205128205128,
|
2083 |
+
"grad_norm": 34.885948181152344,
|
2084 |
+
"learning_rate": 8.6113e-06,
|
2085 |
+
"loss": 1.0293,
|
2086 |
+
"step": 13900
|
2087 |
+
},
|
2088 |
+
{
|
2089 |
+
"epoch": 71.28205128205128,
|
2090 |
+
"eval_loss": 1.1347354650497437,
|
2091 |
+
"eval_runtime": 33.5185,
|
2092 |
+
"eval_samples_per_second": 11.725,
|
2093 |
+
"eval_steps_per_second": 1.492,
|
2094 |
+
"step": 13900
|
2095 |
+
},
|
2096 |
+
{
|
2097 |
+
"epoch": 71.7948717948718,
|
2098 |
+
"grad_norm": 53.24892807006836,
|
2099 |
+
"learning_rate": 8.6013e-06,
|
2100 |
+
"loss": 1.0221,
|
2101 |
+
"step": 14000
|
2102 |
+
},
|
2103 |
+
{
|
2104 |
+
"epoch": 71.7948717948718,
|
2105 |
+
"eval_loss": 1.1401786804199219,
|
2106 |
+
"eval_runtime": 33.3943,
|
2107 |
+
"eval_samples_per_second": 11.768,
|
2108 |
+
"eval_steps_per_second": 1.497,
|
2109 |
+
"step": 14000
|
2110 |
+
},
|
2111 |
+
{
|
2112 |
+
"epoch": 72.3076923076923,
|
2113 |
+
"grad_norm": 19.6131591796875,
|
2114 |
+
"learning_rate": 8.5913e-06,
|
2115 |
+
"loss": 1.022,
|
2116 |
+
"step": 14100
|
2117 |
+
},
|
2118 |
+
{
|
2119 |
+
"epoch": 72.3076923076923,
|
2120 |
+
"eval_loss": 1.129537582397461,
|
2121 |
+
"eval_runtime": 33.4629,
|
2122 |
+
"eval_samples_per_second": 11.744,
|
2123 |
+
"eval_steps_per_second": 1.494,
|
2124 |
+
"step": 14100
|
2125 |
+
},
|
2126 |
+
{
|
2127 |
+
"epoch": 72.82051282051282,
|
2128 |
+
"grad_norm": 25.63981056213379,
|
2129 |
+
"learning_rate": 8.5813e-06,
|
2130 |
+
"loss": 1.0384,
|
2131 |
+
"step": 14200
|
2132 |
+
},
|
2133 |
+
{
|
2134 |
+
"epoch": 72.82051282051282,
|
2135 |
+
"eval_loss": 1.144016981124878,
|
2136 |
+
"eval_runtime": 33.3781,
|
2137 |
+
"eval_samples_per_second": 11.774,
|
2138 |
+
"eval_steps_per_second": 1.498,
|
2139 |
+
"step": 14200
|
2140 |
+
},
|
2141 |
+
{
|
2142 |
+
"epoch": 73.33333333333333,
|
2143 |
+
"grad_norm": 51.96178436279297,
|
2144 |
+
"learning_rate": 8.5713e-06,
|
2145 |
+
"loss": 1.0207,
|
2146 |
+
"step": 14300
|
2147 |
+
},
|
2148 |
+
{
|
2149 |
+
"epoch": 73.33333333333333,
|
2150 |
+
"eval_loss": 1.1289783716201782,
|
2151 |
+
"eval_runtime": 33.3506,
|
2152 |
+
"eval_samples_per_second": 11.784,
|
2153 |
+
"eval_steps_per_second": 1.499,
|
2154 |
+
"step": 14300
|
2155 |
+
},
|
2156 |
+
{
|
2157 |
+
"epoch": 73.84615384615384,
|
2158 |
+
"grad_norm": 72.21915435791016,
|
2159 |
+
"learning_rate": 8.5613e-06,
|
2160 |
+
"loss": 1.0069,
|
2161 |
+
"step": 14400
|
2162 |
+
},
|
2163 |
+
{
|
2164 |
+
"epoch": 73.84615384615384,
|
2165 |
+
"eval_loss": 1.1414133310317993,
|
2166 |
+
"eval_runtime": 33.8899,
|
2167 |
+
"eval_samples_per_second": 11.596,
|
2168 |
+
"eval_steps_per_second": 1.475,
|
2169 |
+
"step": 14400
|
2170 |
+
},
|
2171 |
+
{
|
2172 |
+
"epoch": 74.35897435897436,
|
2173 |
+
"grad_norm": 29.42729949951172,
|
2174 |
+
"learning_rate": 8.5513e-06,
|
2175 |
+
"loss": 1.0113,
|
2176 |
+
"step": 14500
|
2177 |
+
},
|
2178 |
+
{
|
2179 |
+
"epoch": 74.35897435897436,
|
2180 |
+
"eval_loss": 1.1373423337936401,
|
2181 |
+
"eval_runtime": 33.512,
|
2182 |
+
"eval_samples_per_second": 11.727,
|
2183 |
+
"eval_steps_per_second": 1.492,
|
2184 |
+
"step": 14500
|
2185 |
+
},
|
2186 |
+
{
|
2187 |
+
"epoch": 74.87179487179488,
|
2188 |
+
"grad_norm": 26.407670974731445,
|
2189 |
+
"learning_rate": 8.5413e-06,
|
2190 |
+
"loss": 0.9936,
|
2191 |
+
"step": 14600
|
2192 |
+
},
|
2193 |
+
{
|
2194 |
+
"epoch": 74.87179487179488,
|
2195 |
+
"eval_loss": 1.1256376504898071,
|
2196 |
+
"eval_runtime": 33.5297,
|
2197 |
+
"eval_samples_per_second": 11.721,
|
2198 |
+
"eval_steps_per_second": 1.491,
|
2199 |
+
"step": 14600
|
2200 |
+
},
|
2201 |
+
{
|
2202 |
+
"epoch": 75.38461538461539,
|
2203 |
+
"grad_norm": 38.80107498168945,
|
2204 |
+
"learning_rate": 8.5313e-06,
|
2205 |
+
"loss": 1.0198,
|
2206 |
+
"step": 14700
|
2207 |
+
},
|
2208 |
+
{
|
2209 |
+
"epoch": 75.38461538461539,
|
2210 |
+
"eval_loss": 1.1252892017364502,
|
2211 |
+
"eval_runtime": 33.5292,
|
2212 |
+
"eval_samples_per_second": 11.721,
|
2213 |
+
"eval_steps_per_second": 1.491,
|
2214 |
+
"step": 14700
|
2215 |
+
},
|
2216 |
+
{
|
2217 |
+
"epoch": 75.8974358974359,
|
2218 |
+
"grad_norm": 34.42100524902344,
|
2219 |
+
"learning_rate": 8.521300000000001e-06,
|
2220 |
+
"loss": 1.0124,
|
2221 |
+
"step": 14800
|
2222 |
+
},
|
2223 |
+
{
|
2224 |
+
"epoch": 75.8974358974359,
|
2225 |
+
"eval_loss": 1.1165941953659058,
|
2226 |
+
"eval_runtime": 33.3962,
|
2227 |
+
"eval_samples_per_second": 11.768,
|
2228 |
+
"eval_steps_per_second": 1.497,
|
2229 |
+
"step": 14800
|
2230 |
+
},
|
2231 |
+
{
|
2232 |
+
"epoch": 76.41025641025641,
|
2233 |
+
"grad_norm": 24.1181697845459,
|
2234 |
+
"learning_rate": 8.511300000000001e-06,
|
2235 |
+
"loss": 1.0331,
|
2236 |
+
"step": 14900
|
2237 |
+
},
|
2238 |
+
{
|
2239 |
+
"epoch": 76.41025641025641,
|
2240 |
+
"eval_loss": 1.1260528564453125,
|
2241 |
+
"eval_runtime": 33.6091,
|
2242 |
+
"eval_samples_per_second": 11.693,
|
2243 |
+
"eval_steps_per_second": 1.488,
|
2244 |
+
"step": 14900
|
2245 |
+
},
|
2246 |
+
{
|
2247 |
+
"epoch": 76.92307692307692,
|
2248 |
+
"grad_norm": 31.163135528564453,
|
2249 |
+
"learning_rate": 8.501300000000001e-06,
|
2250 |
+
"loss": 0.9759,
|
2251 |
+
"step": 15000
|
2252 |
+
},
|
2253 |
+
{
|
2254 |
+
"epoch": 76.92307692307692,
|
2255 |
+
"eval_loss": 1.1240051984786987,
|
2256 |
+
"eval_runtime": 33.4198,
|
2257 |
+
"eval_samples_per_second": 11.759,
|
2258 |
+
"eval_steps_per_second": 1.496,
|
2259 |
+
"step": 15000
|
2260 |
+
}
|
2261 |
+
],
|
2262 |
+
"logging_steps": 100,
|
2263 |
+
"max_steps": 100000,
|
2264 |
+
"num_input_tokens_seen": 0,
|
2265 |
+
"num_train_epochs": 513,
|
2266 |
+
"save_steps": 500,
|
2267 |
+
"stateful_callbacks": {
|
2268 |
+
"TrainerControl": {
|
2269 |
+
"args": {
|
2270 |
+
"should_epoch_stop": false,
|
2271 |
+
"should_evaluate": false,
|
2272 |
+
"should_log": false,
|
2273 |
+
"should_save": true,
|
2274 |
+
"should_training_stop": false
|
2275 |
+
},
|
2276 |
+
"attributes": {}
|
2277 |
+
}
|
2278 |
+
},
|
2279 |
+
"total_flos": 5.73364749312e+19,
|
2280 |
+
"train_batch_size": 8,
|
2281 |
+
"trial_name": null,
|
2282 |
+
"trial_params": null
|
2283 |
+
}
|
checkpoint-15000/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:12024124d98d6f3527d5a0e9a760923228a8928111a77f0d825458c2cc9072c6
|
3 |
+
size 5304
|