checkpoint-1715
Browse files- config.json +147 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- preprocessor_config.json +23 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +1270 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,147 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "microsoft/swinv2-base-patch4-window16-256",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"Swinv2ForImageClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.0,
|
| 7 |
+
"depths": [
|
| 8 |
+
2,
|
| 9 |
+
2,
|
| 10 |
+
18,
|
| 11 |
+
2
|
| 12 |
+
],
|
| 13 |
+
"drop_path_rate": 0.1,
|
| 14 |
+
"embed_dim": 128,
|
| 15 |
+
"encoder_stride": 32,
|
| 16 |
+
"hidden_act": "gelu",
|
| 17 |
+
"hidden_dropout_prob": 0.0,
|
| 18 |
+
"hidden_size": 1024,
|
| 19 |
+
"id2label": {
|
| 20 |
+
"0": "LABEL_0",
|
| 21 |
+
"1": "LABEL_1",
|
| 22 |
+
"2": "LABEL_2",
|
| 23 |
+
"3": "LABEL_3",
|
| 24 |
+
"4": "LABEL_4",
|
| 25 |
+
"5": "LABEL_5",
|
| 26 |
+
"6": "LABEL_6",
|
| 27 |
+
"7": "LABEL_7",
|
| 28 |
+
"8": "LABEL_8",
|
| 29 |
+
"9": "LABEL_9",
|
| 30 |
+
"10": "LABEL_10",
|
| 31 |
+
"11": "LABEL_11",
|
| 32 |
+
"12": "LABEL_12",
|
| 33 |
+
"13": "LABEL_13",
|
| 34 |
+
"14": "LABEL_14",
|
| 35 |
+
"15": "LABEL_15",
|
| 36 |
+
"16": "LABEL_16",
|
| 37 |
+
"17": "LABEL_17",
|
| 38 |
+
"18": "LABEL_18",
|
| 39 |
+
"19": "LABEL_19",
|
| 40 |
+
"20": "LABEL_20",
|
| 41 |
+
"21": "LABEL_21",
|
| 42 |
+
"22": "LABEL_22",
|
| 43 |
+
"23": "LABEL_23",
|
| 44 |
+
"24": "LABEL_24",
|
| 45 |
+
"25": "LABEL_25",
|
| 46 |
+
"26": "LABEL_26",
|
| 47 |
+
"27": "LABEL_27",
|
| 48 |
+
"28": "LABEL_28",
|
| 49 |
+
"29": "LABEL_29",
|
| 50 |
+
"30": "LABEL_30",
|
| 51 |
+
"31": "LABEL_31",
|
| 52 |
+
"32": "LABEL_32",
|
| 53 |
+
"33": "LABEL_33",
|
| 54 |
+
"34": "LABEL_34",
|
| 55 |
+
"35": "LABEL_35",
|
| 56 |
+
"36": "LABEL_36",
|
| 57 |
+
"37": "LABEL_37",
|
| 58 |
+
"38": "LABEL_38",
|
| 59 |
+
"39": "LABEL_39",
|
| 60 |
+
"40": "LABEL_40",
|
| 61 |
+
"41": "LABEL_41"
|
| 62 |
+
},
|
| 63 |
+
"image_size": 256,
|
| 64 |
+
"initializer_range": 0.02,
|
| 65 |
+
"label2id": {
|
| 66 |
+
"LABEL_0": 0,
|
| 67 |
+
"LABEL_1": 1,
|
| 68 |
+
"LABEL_10": 10,
|
| 69 |
+
"LABEL_11": 11,
|
| 70 |
+
"LABEL_12": 12,
|
| 71 |
+
"LABEL_13": 13,
|
| 72 |
+
"LABEL_14": 14,
|
| 73 |
+
"LABEL_15": 15,
|
| 74 |
+
"LABEL_16": 16,
|
| 75 |
+
"LABEL_17": 17,
|
| 76 |
+
"LABEL_18": 18,
|
| 77 |
+
"LABEL_19": 19,
|
| 78 |
+
"LABEL_2": 2,
|
| 79 |
+
"LABEL_20": 20,
|
| 80 |
+
"LABEL_21": 21,
|
| 81 |
+
"LABEL_22": 22,
|
| 82 |
+
"LABEL_23": 23,
|
| 83 |
+
"LABEL_24": 24,
|
| 84 |
+
"LABEL_25": 25,
|
| 85 |
+
"LABEL_26": 26,
|
| 86 |
+
"LABEL_27": 27,
|
| 87 |
+
"LABEL_28": 28,
|
| 88 |
+
"LABEL_29": 29,
|
| 89 |
+
"LABEL_3": 3,
|
| 90 |
+
"LABEL_30": 30,
|
| 91 |
+
"LABEL_31": 31,
|
| 92 |
+
"LABEL_32": 32,
|
| 93 |
+
"LABEL_33": 33,
|
| 94 |
+
"LABEL_34": 34,
|
| 95 |
+
"LABEL_35": 35,
|
| 96 |
+
"LABEL_36": 36,
|
| 97 |
+
"LABEL_37": 37,
|
| 98 |
+
"LABEL_38": 38,
|
| 99 |
+
"LABEL_39": 39,
|
| 100 |
+
"LABEL_4": 4,
|
| 101 |
+
"LABEL_40": 40,
|
| 102 |
+
"LABEL_41": 41,
|
| 103 |
+
"LABEL_5": 5,
|
| 104 |
+
"LABEL_6": 6,
|
| 105 |
+
"LABEL_7": 7,
|
| 106 |
+
"LABEL_8": 8,
|
| 107 |
+
"LABEL_9": 9
|
| 108 |
+
},
|
| 109 |
+
"layer_norm_eps": 1e-05,
|
| 110 |
+
"mlp_ratio": 4.0,
|
| 111 |
+
"model_type": "swinv2",
|
| 112 |
+
"num_channels": 3,
|
| 113 |
+
"num_heads": [
|
| 114 |
+
4,
|
| 115 |
+
8,
|
| 116 |
+
16,
|
| 117 |
+
32
|
| 118 |
+
],
|
| 119 |
+
"num_layers": 4,
|
| 120 |
+
"out_features": [
|
| 121 |
+
"stage4"
|
| 122 |
+
],
|
| 123 |
+
"out_indices": [
|
| 124 |
+
4
|
| 125 |
+
],
|
| 126 |
+
"patch_size": 4,
|
| 127 |
+
"path_norm": true,
|
| 128 |
+
"pretrained_window_sizes": [
|
| 129 |
+
0,
|
| 130 |
+
0,
|
| 131 |
+
0,
|
| 132 |
+
0
|
| 133 |
+
],
|
| 134 |
+
"problem_type": "single_label_classification",
|
| 135 |
+
"qkv_bias": true,
|
| 136 |
+
"stage_names": [
|
| 137 |
+
"stem",
|
| 138 |
+
"stage1",
|
| 139 |
+
"stage2",
|
| 140 |
+
"stage3",
|
| 141 |
+
"stage4"
|
| 142 |
+
],
|
| 143 |
+
"torch_dtype": "float32",
|
| 144 |
+
"transformers_version": "4.47.0",
|
| 145 |
+
"use_absolute_embeddings": false,
|
| 146 |
+
"window_size": 16
|
| 147 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f3ebf02c412843b8598e0c0cc38f1e9573d5664dcf2e89a33fff0bbf0de739a7
|
| 3 |
+
size 347809528
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4672721aeaed7851b718719ba89d07b8faade79816ada69b742235b791c02554
|
| 3 |
+
size 695890973
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_convert_rgb": null,
|
| 3 |
+
"do_normalize": true,
|
| 4 |
+
"do_rescale": true,
|
| 5 |
+
"do_resize": true,
|
| 6 |
+
"image_mean": [
|
| 7 |
+
0.485,
|
| 8 |
+
0.456,
|
| 9 |
+
0.406
|
| 10 |
+
],
|
| 11 |
+
"image_processor_type": "ViTImageProcessor",
|
| 12 |
+
"image_std": [
|
| 13 |
+
0.229,
|
| 14 |
+
0.224,
|
| 15 |
+
0.225
|
| 16 |
+
],
|
| 17 |
+
"resample": 3,
|
| 18 |
+
"rescale_factor": 0.00392156862745098,
|
| 19 |
+
"size": {
|
| 20 |
+
"height": 256,
|
| 21 |
+
"width": 256
|
| 22 |
+
}
|
| 23 |
+
}
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:92cc70c170169af5d978185b92767499e956a5f46adbffc2bd988ec8371ec7de
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:949b2b0a0f1ad791a5c9823d31ee47a7380d3468ca4b2ac70a9908fee83c9a58
|
| 3 |
+
size 1064
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1270 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1715,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.029154518950437316,
|
| 13 |
+
"grad_norm": 6.396158218383789,
|
| 14 |
+
"learning_rate": 4.970845481049563e-05,
|
| 15 |
+
"loss": 3.6944,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.05830903790087463,
|
| 20 |
+
"grad_norm": 5.8870744705200195,
|
| 21 |
+
"learning_rate": 4.941690962099126e-05,
|
| 22 |
+
"loss": 3.4646,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.08746355685131195,
|
| 27 |
+
"grad_norm": 8.024749755859375,
|
| 28 |
+
"learning_rate": 4.912536443148688e-05,
|
| 29 |
+
"loss": 3.2603,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.11661807580174927,
|
| 34 |
+
"grad_norm": 12.580345153808594,
|
| 35 |
+
"learning_rate": 4.883381924198251e-05,
|
| 36 |
+
"loss": 2.9375,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.1457725947521866,
|
| 41 |
+
"grad_norm": 13.44204330444336,
|
| 42 |
+
"learning_rate": 4.8542274052478136e-05,
|
| 43 |
+
"loss": 2.663,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.1749271137026239,
|
| 48 |
+
"grad_norm": 13.08288288116455,
|
| 49 |
+
"learning_rate": 4.825072886297377e-05,
|
| 50 |
+
"loss": 2.4086,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.20408163265306123,
|
| 55 |
+
"grad_norm": 14.201903343200684,
|
| 56 |
+
"learning_rate": 4.795918367346939e-05,
|
| 57 |
+
"loss": 2.0242,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.23323615160349853,
|
| 62 |
+
"grad_norm": 14.91315746307373,
|
| 63 |
+
"learning_rate": 4.7667638483965015e-05,
|
| 64 |
+
"loss": 1.8438,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.26239067055393583,
|
| 69 |
+
"grad_norm": 14.440930366516113,
|
| 70 |
+
"learning_rate": 4.7376093294460646e-05,
|
| 71 |
+
"loss": 1.7379,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.2915451895043732,
|
| 76 |
+
"grad_norm": 14.924482345581055,
|
| 77 |
+
"learning_rate": 4.708454810495627e-05,
|
| 78 |
+
"loss": 1.3515,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.3206997084548105,
|
| 83 |
+
"grad_norm": 15.159282684326172,
|
| 84 |
+
"learning_rate": 4.6793002915451894e-05,
|
| 85 |
+
"loss": 1.4493,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.3498542274052478,
|
| 90 |
+
"grad_norm": 15.176348686218262,
|
| 91 |
+
"learning_rate": 4.6501457725947525e-05,
|
| 92 |
+
"loss": 1.3174,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.37900874635568516,
|
| 97 |
+
"grad_norm": 15.329106330871582,
|
| 98 |
+
"learning_rate": 4.620991253644315e-05,
|
| 99 |
+
"loss": 1.3243,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.40816326530612246,
|
| 104 |
+
"grad_norm": 13.873916625976562,
|
| 105 |
+
"learning_rate": 4.591836734693878e-05,
|
| 106 |
+
"loss": 1.0203,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.43731778425655976,
|
| 111 |
+
"grad_norm": 11.680057525634766,
|
| 112 |
+
"learning_rate": 4.5626822157434404e-05,
|
| 113 |
+
"loss": 1.2277,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.46647230320699706,
|
| 118 |
+
"grad_norm": 17.877845764160156,
|
| 119 |
+
"learning_rate": 4.533527696793003e-05,
|
| 120 |
+
"loss": 0.9823,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.4956268221574344,
|
| 125 |
+
"grad_norm": 12.79694938659668,
|
| 126 |
+
"learning_rate": 4.504373177842566e-05,
|
| 127 |
+
"loss": 1.1087,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.5247813411078717,
|
| 132 |
+
"grad_norm": 15.178705215454102,
|
| 133 |
+
"learning_rate": 4.475218658892128e-05,
|
| 134 |
+
"loss": 0.9267,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.5539358600583091,
|
| 139 |
+
"grad_norm": 19.371179580688477,
|
| 140 |
+
"learning_rate": 4.4460641399416914e-05,
|
| 141 |
+
"loss": 1.1024,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.5830903790087464,
|
| 146 |
+
"grad_norm": 12.616296768188477,
|
| 147 |
+
"learning_rate": 4.416909620991254e-05,
|
| 148 |
+
"loss": 0.9355,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.6122448979591837,
|
| 153 |
+
"grad_norm": 15.549710273742676,
|
| 154 |
+
"learning_rate": 4.387755102040816e-05,
|
| 155 |
+
"loss": 0.8048,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.641399416909621,
|
| 160 |
+
"grad_norm": 15.638792037963867,
|
| 161 |
+
"learning_rate": 4.358600583090379e-05,
|
| 162 |
+
"loss": 0.8175,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.6705539358600583,
|
| 167 |
+
"grad_norm": 12.656390190124512,
|
| 168 |
+
"learning_rate": 4.3294460641399424e-05,
|
| 169 |
+
"loss": 1.0265,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.6997084548104956,
|
| 174 |
+
"grad_norm": 12.056506156921387,
|
| 175 |
+
"learning_rate": 4.300291545189505e-05,
|
| 176 |
+
"loss": 0.8383,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.7288629737609329,
|
| 181 |
+
"grad_norm": 13.179421424865723,
|
| 182 |
+
"learning_rate": 4.271137026239067e-05,
|
| 183 |
+
"loss": 0.882,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.7580174927113703,
|
| 188 |
+
"grad_norm": 11.642322540283203,
|
| 189 |
+
"learning_rate": 4.2419825072886296e-05,
|
| 190 |
+
"loss": 0.9375,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.7871720116618076,
|
| 195 |
+
"grad_norm": 14.77869987487793,
|
| 196 |
+
"learning_rate": 4.212827988338193e-05,
|
| 197 |
+
"loss": 0.721,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.8163265306122449,
|
| 202 |
+
"grad_norm": 12.768646240234375,
|
| 203 |
+
"learning_rate": 4.183673469387756e-05,
|
| 204 |
+
"loss": 0.7863,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.8454810495626822,
|
| 209 |
+
"grad_norm": 17.440149307250977,
|
| 210 |
+
"learning_rate": 4.1545189504373175e-05,
|
| 211 |
+
"loss": 0.7854,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.8746355685131195,
|
| 216 |
+
"grad_norm": 14.728607177734375,
|
| 217 |
+
"learning_rate": 4.1253644314868806e-05,
|
| 218 |
+
"loss": 0.6795,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.9037900874635568,
|
| 223 |
+
"grad_norm": 13.96823787689209,
|
| 224 |
+
"learning_rate": 4.0962099125364436e-05,
|
| 225 |
+
"loss": 0.8244,
|
| 226 |
+
"step": 310
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.9329446064139941,
|
| 230 |
+
"grad_norm": 14.88004207611084,
|
| 231 |
+
"learning_rate": 4.067055393586006e-05,
|
| 232 |
+
"loss": 0.8082,
|
| 233 |
+
"step": 320
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.9620991253644315,
|
| 237 |
+
"grad_norm": 12.001025199890137,
|
| 238 |
+
"learning_rate": 4.0379008746355685e-05,
|
| 239 |
+
"loss": 0.7094,
|
| 240 |
+
"step": 330
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.9912536443148688,
|
| 244 |
+
"grad_norm": 20.970056533813477,
|
| 245 |
+
"learning_rate": 4.0087463556851315e-05,
|
| 246 |
+
"loss": 0.8157,
|
| 247 |
+
"step": 340
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 1.0,
|
| 251 |
+
"eval_loss": 0.7410290837287903,
|
| 252 |
+
"eval_runtime": 346.7173,
|
| 253 |
+
"eval_samples_per_second": 7.888,
|
| 254 |
+
"eval_steps_per_second": 0.248,
|
| 255 |
+
"step": 343
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 1.0204081632653061,
|
| 259 |
+
"grad_norm": 8.203807830810547,
|
| 260 |
+
"learning_rate": 3.979591836734694e-05,
|
| 261 |
+
"loss": 0.4442,
|
| 262 |
+
"step": 350
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 1.0495626822157433,
|
| 266 |
+
"grad_norm": 9.512131690979004,
|
| 267 |
+
"learning_rate": 3.950437317784257e-05,
|
| 268 |
+
"loss": 0.4501,
|
| 269 |
+
"step": 360
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 1.0787172011661808,
|
| 273 |
+
"grad_norm": 11.348304748535156,
|
| 274 |
+
"learning_rate": 3.9212827988338194e-05,
|
| 275 |
+
"loss": 0.5578,
|
| 276 |
+
"step": 370
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 1.1078717201166182,
|
| 280 |
+
"grad_norm": 16.04487419128418,
|
| 281 |
+
"learning_rate": 3.892128279883382e-05,
|
| 282 |
+
"loss": 0.5277,
|
| 283 |
+
"step": 380
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 1.1370262390670554,
|
| 287 |
+
"grad_norm": 11.712241172790527,
|
| 288 |
+
"learning_rate": 3.862973760932945e-05,
|
| 289 |
+
"loss": 0.4313,
|
| 290 |
+
"step": 390
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 1.1661807580174928,
|
| 294 |
+
"grad_norm": 10.177406311035156,
|
| 295 |
+
"learning_rate": 3.833819241982507e-05,
|
| 296 |
+
"loss": 0.4427,
|
| 297 |
+
"step": 400
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 1.19533527696793,
|
| 301 |
+
"grad_norm": 10.864590644836426,
|
| 302 |
+
"learning_rate": 3.8046647230320704e-05,
|
| 303 |
+
"loss": 0.4545,
|
| 304 |
+
"step": 410
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 1.2244897959183674,
|
| 308 |
+
"grad_norm": 27.668434143066406,
|
| 309 |
+
"learning_rate": 3.775510204081633e-05,
|
| 310 |
+
"loss": 0.3945,
|
| 311 |
+
"step": 420
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 1.2536443148688048,
|
| 315 |
+
"grad_norm": 12.273193359375,
|
| 316 |
+
"learning_rate": 3.746355685131195e-05,
|
| 317 |
+
"loss": 0.4442,
|
| 318 |
+
"step": 430
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 1.282798833819242,
|
| 322 |
+
"grad_norm": 10.925050735473633,
|
| 323 |
+
"learning_rate": 3.717201166180758e-05,
|
| 324 |
+
"loss": 0.3749,
|
| 325 |
+
"step": 440
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 1.3119533527696792,
|
| 329 |
+
"grad_norm": 15.893428802490234,
|
| 330 |
+
"learning_rate": 3.688046647230321e-05,
|
| 331 |
+
"loss": 0.3903,
|
| 332 |
+
"step": 450
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 1.3411078717201166,
|
| 336 |
+
"grad_norm": 17.293563842773438,
|
| 337 |
+
"learning_rate": 3.658892128279884e-05,
|
| 338 |
+
"loss": 0.5249,
|
| 339 |
+
"step": 460
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 1.370262390670554,
|
| 343 |
+
"grad_norm": 10.63139820098877,
|
| 344 |
+
"learning_rate": 3.629737609329446e-05,
|
| 345 |
+
"loss": 0.4097,
|
| 346 |
+
"step": 470
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 1.3994169096209912,
|
| 350 |
+
"grad_norm": 10.485889434814453,
|
| 351 |
+
"learning_rate": 3.6005830903790086e-05,
|
| 352 |
+
"loss": 0.4057,
|
| 353 |
+
"step": 480
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 1.4285714285714286,
|
| 357 |
+
"grad_norm": 15.918136596679688,
|
| 358 |
+
"learning_rate": 3.571428571428572e-05,
|
| 359 |
+
"loss": 0.333,
|
| 360 |
+
"step": 490
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.4577259475218658,
|
| 364 |
+
"grad_norm": 5.628406047821045,
|
| 365 |
+
"learning_rate": 3.542274052478135e-05,
|
| 366 |
+
"loss": 0.4202,
|
| 367 |
+
"step": 500
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 1.4868804664723032,
|
| 371 |
+
"grad_norm": 8.147640228271484,
|
| 372 |
+
"learning_rate": 3.5131195335276965e-05,
|
| 373 |
+
"loss": 0.3755,
|
| 374 |
+
"step": 510
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 1.5160349854227406,
|
| 378 |
+
"grad_norm": 11.122663497924805,
|
| 379 |
+
"learning_rate": 3.4839650145772596e-05,
|
| 380 |
+
"loss": 0.3533,
|
| 381 |
+
"step": 520
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 1.5451895043731778,
|
| 385 |
+
"grad_norm": 8.978357315063477,
|
| 386 |
+
"learning_rate": 3.454810495626823e-05,
|
| 387 |
+
"loss": 0.3938,
|
| 388 |
+
"step": 530
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.574344023323615,
|
| 392 |
+
"grad_norm": 13.574569702148438,
|
| 393 |
+
"learning_rate": 3.425655976676385e-05,
|
| 394 |
+
"loss": 0.4004,
|
| 395 |
+
"step": 540
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 1.6034985422740524,
|
| 399 |
+
"grad_norm": 12.93692398071289,
|
| 400 |
+
"learning_rate": 3.3965014577259475e-05,
|
| 401 |
+
"loss": 0.3081,
|
| 402 |
+
"step": 550
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 1.6326530612244898,
|
| 406 |
+
"grad_norm": 9.078178405761719,
|
| 407 |
+
"learning_rate": 3.36734693877551e-05,
|
| 408 |
+
"loss": 0.3557,
|
| 409 |
+
"step": 560
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 1.6618075801749272,
|
| 413 |
+
"grad_norm": 7.011285305023193,
|
| 414 |
+
"learning_rate": 3.338192419825073e-05,
|
| 415 |
+
"loss": 0.3968,
|
| 416 |
+
"step": 570
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 1.6909620991253644,
|
| 420 |
+
"grad_norm": 12.261427879333496,
|
| 421 |
+
"learning_rate": 3.309037900874636e-05,
|
| 422 |
+
"loss": 0.349,
|
| 423 |
+
"step": 580
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 1.7201166180758016,
|
| 427 |
+
"grad_norm": 7.620883464813232,
|
| 428 |
+
"learning_rate": 3.2798833819241985e-05,
|
| 429 |
+
"loss": 0.3861,
|
| 430 |
+
"step": 590
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 1.749271137026239,
|
| 434 |
+
"grad_norm": 7.384204864501953,
|
| 435 |
+
"learning_rate": 3.250728862973761e-05,
|
| 436 |
+
"loss": 0.3951,
|
| 437 |
+
"step": 600
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 1.7784256559766765,
|
| 441 |
+
"grad_norm": 11.785451889038086,
|
| 442 |
+
"learning_rate": 3.221574344023324e-05,
|
| 443 |
+
"loss": 0.4344,
|
| 444 |
+
"step": 610
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 1.8075801749271136,
|
| 448 |
+
"grad_norm": 9.102931022644043,
|
| 449 |
+
"learning_rate": 3.1924198250728864e-05,
|
| 450 |
+
"loss": 0.3569,
|
| 451 |
+
"step": 620
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 1.836734693877551,
|
| 455 |
+
"grad_norm": 12.737972259521484,
|
| 456 |
+
"learning_rate": 3.1632653061224494e-05,
|
| 457 |
+
"loss": 0.4627,
|
| 458 |
+
"step": 630
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 1.8658892128279883,
|
| 462 |
+
"grad_norm": 6.976524829864502,
|
| 463 |
+
"learning_rate": 3.134110787172012e-05,
|
| 464 |
+
"loss": 0.4338,
|
| 465 |
+
"step": 640
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 1.8950437317784257,
|
| 469 |
+
"grad_norm": 10.488119125366211,
|
| 470 |
+
"learning_rate": 3.104956268221574e-05,
|
| 471 |
+
"loss": 0.4369,
|
| 472 |
+
"step": 650
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 1.924198250728863,
|
| 476 |
+
"grad_norm": 13.129137992858887,
|
| 477 |
+
"learning_rate": 3.0758017492711373e-05,
|
| 478 |
+
"loss": 0.4023,
|
| 479 |
+
"step": 660
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 1.9533527696793003,
|
| 483 |
+
"grad_norm": 11.848146438598633,
|
| 484 |
+
"learning_rate": 3.0466472303207e-05,
|
| 485 |
+
"loss": 0.3211,
|
| 486 |
+
"step": 670
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 1.9825072886297375,
|
| 490 |
+
"grad_norm": 13.786744117736816,
|
| 491 |
+
"learning_rate": 3.017492711370263e-05,
|
| 492 |
+
"loss": 0.4545,
|
| 493 |
+
"step": 680
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 2.0,
|
| 497 |
+
"eval_loss": 0.6418492794036865,
|
| 498 |
+
"eval_runtime": 344.1102,
|
| 499 |
+
"eval_samples_per_second": 7.948,
|
| 500 |
+
"eval_steps_per_second": 0.25,
|
| 501 |
+
"step": 686
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 2.011661807580175,
|
| 505 |
+
"grad_norm": 7.113315582275391,
|
| 506 |
+
"learning_rate": 2.988338192419825e-05,
|
| 507 |
+
"loss": 0.3137,
|
| 508 |
+
"step": 690
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 2.0408163265306123,
|
| 512 |
+
"grad_norm": 9.555095672607422,
|
| 513 |
+
"learning_rate": 2.959183673469388e-05,
|
| 514 |
+
"loss": 0.1409,
|
| 515 |
+
"step": 700
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 2.0699708454810497,
|
| 519 |
+
"grad_norm": 14.719175338745117,
|
| 520 |
+
"learning_rate": 2.9300291545189507e-05,
|
| 521 |
+
"loss": 0.1807,
|
| 522 |
+
"step": 710
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 2.0991253644314867,
|
| 526 |
+
"grad_norm": 6.63716459274292,
|
| 527 |
+
"learning_rate": 2.9008746355685135e-05,
|
| 528 |
+
"loss": 0.1116,
|
| 529 |
+
"step": 720
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 2.128279883381924,
|
| 533 |
+
"grad_norm": 4.129220962524414,
|
| 534 |
+
"learning_rate": 2.871720116618076e-05,
|
| 535 |
+
"loss": 0.1921,
|
| 536 |
+
"step": 730
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 2.1574344023323615,
|
| 540 |
+
"grad_norm": 7.979690074920654,
|
| 541 |
+
"learning_rate": 2.8425655976676386e-05,
|
| 542 |
+
"loss": 0.1378,
|
| 543 |
+
"step": 740
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 2.186588921282799,
|
| 547 |
+
"grad_norm": 11.688068389892578,
|
| 548 |
+
"learning_rate": 2.8134110787172014e-05,
|
| 549 |
+
"loss": 0.1304,
|
| 550 |
+
"step": 750
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 2.2157434402332363,
|
| 554 |
+
"grad_norm": 9.168377876281738,
|
| 555 |
+
"learning_rate": 2.784256559766764e-05,
|
| 556 |
+
"loss": 0.1774,
|
| 557 |
+
"step": 760
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 2.2448979591836733,
|
| 561 |
+
"grad_norm": 14.069756507873535,
|
| 562 |
+
"learning_rate": 2.7551020408163265e-05,
|
| 563 |
+
"loss": 0.1482,
|
| 564 |
+
"step": 770
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 2.2740524781341107,
|
| 568 |
+
"grad_norm": 2.5552010536193848,
|
| 569 |
+
"learning_rate": 2.7259475218658893e-05,
|
| 570 |
+
"loss": 0.1194,
|
| 571 |
+
"step": 780
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 2.303206997084548,
|
| 575 |
+
"grad_norm": 7.534204959869385,
|
| 576 |
+
"learning_rate": 2.696793002915452e-05,
|
| 577 |
+
"loss": 0.1321,
|
| 578 |
+
"step": 790
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 2.3323615160349855,
|
| 582 |
+
"grad_norm": 6.635725021362305,
|
| 583 |
+
"learning_rate": 2.6676384839650148e-05,
|
| 584 |
+
"loss": 0.0902,
|
| 585 |
+
"step": 800
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 2.3615160349854225,
|
| 589 |
+
"grad_norm": 4.213601589202881,
|
| 590 |
+
"learning_rate": 2.6384839650145775e-05,
|
| 591 |
+
"loss": 0.182,
|
| 592 |
+
"step": 810
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 2.39067055393586,
|
| 596 |
+
"grad_norm": 11.590983390808105,
|
| 597 |
+
"learning_rate": 2.60932944606414e-05,
|
| 598 |
+
"loss": 0.1745,
|
| 599 |
+
"step": 820
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 2.4198250728862973,
|
| 603 |
+
"grad_norm": 9.884819984436035,
|
| 604 |
+
"learning_rate": 2.5801749271137027e-05,
|
| 605 |
+
"loss": 0.1733,
|
| 606 |
+
"step": 830
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 2.4489795918367347,
|
| 610 |
+
"grad_norm": 8.458849906921387,
|
| 611 |
+
"learning_rate": 2.5510204081632654e-05,
|
| 612 |
+
"loss": 0.1307,
|
| 613 |
+
"step": 840
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 2.478134110787172,
|
| 617 |
+
"grad_norm": 7.574510097503662,
|
| 618 |
+
"learning_rate": 2.521865889212828e-05,
|
| 619 |
+
"loss": 0.1253,
|
| 620 |
+
"step": 850
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 2.5072886297376096,
|
| 624 |
+
"grad_norm": 9.187530517578125,
|
| 625 |
+
"learning_rate": 2.492711370262391e-05,
|
| 626 |
+
"loss": 0.1538,
|
| 627 |
+
"step": 860
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 2.5364431486880465,
|
| 631 |
+
"grad_norm": 5.005228519439697,
|
| 632 |
+
"learning_rate": 2.4635568513119533e-05,
|
| 633 |
+
"loss": 0.117,
|
| 634 |
+
"step": 870
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 2.565597667638484,
|
| 638 |
+
"grad_norm": 3.9266200065612793,
|
| 639 |
+
"learning_rate": 2.434402332361516e-05,
|
| 640 |
+
"loss": 0.1449,
|
| 641 |
+
"step": 880
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 2.5947521865889214,
|
| 645 |
+
"grad_norm": 8.087149620056152,
|
| 646 |
+
"learning_rate": 2.405247813411079e-05,
|
| 647 |
+
"loss": 0.1652,
|
| 648 |
+
"step": 890
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 2.6239067055393583,
|
| 652 |
+
"grad_norm": 13.278199195861816,
|
| 653 |
+
"learning_rate": 2.3760932944606415e-05,
|
| 654 |
+
"loss": 0.1834,
|
| 655 |
+
"step": 900
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 2.6530612244897958,
|
| 659 |
+
"grad_norm": 9.696410179138184,
|
| 660 |
+
"learning_rate": 2.3469387755102043e-05,
|
| 661 |
+
"loss": 0.1033,
|
| 662 |
+
"step": 910
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 2.682215743440233,
|
| 666 |
+
"grad_norm": 14.884827613830566,
|
| 667 |
+
"learning_rate": 2.3177842565597667e-05,
|
| 668 |
+
"loss": 0.2509,
|
| 669 |
+
"step": 920
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 2.7113702623906706,
|
| 673 |
+
"grad_norm": 7.274765968322754,
|
| 674 |
+
"learning_rate": 2.2886297376093298e-05,
|
| 675 |
+
"loss": 0.1497,
|
| 676 |
+
"step": 930
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 2.740524781341108,
|
| 680 |
+
"grad_norm": 11.48387622833252,
|
| 681 |
+
"learning_rate": 2.2594752186588922e-05,
|
| 682 |
+
"loss": 0.1106,
|
| 683 |
+
"step": 940
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 2.7696793002915454,
|
| 687 |
+
"grad_norm": 7.414773941040039,
|
| 688 |
+
"learning_rate": 2.230320699708455e-05,
|
| 689 |
+
"loss": 0.167,
|
| 690 |
+
"step": 950
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 2.7988338192419824,
|
| 694 |
+
"grad_norm": 11.676288604736328,
|
| 695 |
+
"learning_rate": 2.2011661807580177e-05,
|
| 696 |
+
"loss": 0.1707,
|
| 697 |
+
"step": 960
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 2.82798833819242,
|
| 701 |
+
"grad_norm": 7.887070178985596,
|
| 702 |
+
"learning_rate": 2.1720116618075804e-05,
|
| 703 |
+
"loss": 0.1284,
|
| 704 |
+
"step": 970
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 2.857142857142857,
|
| 708 |
+
"grad_norm": 16.793603897094727,
|
| 709 |
+
"learning_rate": 2.1428571428571428e-05,
|
| 710 |
+
"loss": 0.1464,
|
| 711 |
+
"step": 980
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 2.8862973760932946,
|
| 715 |
+
"grad_norm": 11.829538345336914,
|
| 716 |
+
"learning_rate": 2.1137026239067056e-05,
|
| 717 |
+
"loss": 0.109,
|
| 718 |
+
"step": 990
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 2.9154518950437316,
|
| 722 |
+
"grad_norm": 6.482255458831787,
|
| 723 |
+
"learning_rate": 2.0845481049562683e-05,
|
| 724 |
+
"loss": 0.1162,
|
| 725 |
+
"step": 1000
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 2.944606413994169,
|
| 729 |
+
"grad_norm": 10.546632766723633,
|
| 730 |
+
"learning_rate": 2.055393586005831e-05,
|
| 731 |
+
"loss": 0.1181,
|
| 732 |
+
"step": 1010
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 2.9737609329446064,
|
| 736 |
+
"grad_norm": 3.5399889945983887,
|
| 737 |
+
"learning_rate": 2.0262390670553938e-05,
|
| 738 |
+
"loss": 0.1005,
|
| 739 |
+
"step": 1020
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 3.0,
|
| 743 |
+
"eval_loss": 0.680114209651947,
|
| 744 |
+
"eval_runtime": 335.8694,
|
| 745 |
+
"eval_samples_per_second": 8.143,
|
| 746 |
+
"eval_steps_per_second": 0.256,
|
| 747 |
+
"step": 1029
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 3.002915451895044,
|
| 751 |
+
"grad_norm": 3.2238965034484863,
|
| 752 |
+
"learning_rate": 1.9970845481049562e-05,
|
| 753 |
+
"loss": 0.166,
|
| 754 |
+
"step": 1030
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 3.0320699708454812,
|
| 758 |
+
"grad_norm": 2.8589000701904297,
|
| 759 |
+
"learning_rate": 1.9679300291545193e-05,
|
| 760 |
+
"loss": 0.0457,
|
| 761 |
+
"step": 1040
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 3.061224489795918,
|
| 765 |
+
"grad_norm": 2.0926849842071533,
|
| 766 |
+
"learning_rate": 1.9387755102040817e-05,
|
| 767 |
+
"loss": 0.0497,
|
| 768 |
+
"step": 1050
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 3.0903790087463556,
|
| 772 |
+
"grad_norm": 10.41883659362793,
|
| 773 |
+
"learning_rate": 1.9096209912536444e-05,
|
| 774 |
+
"loss": 0.0426,
|
| 775 |
+
"step": 1060
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 3.119533527696793,
|
| 779 |
+
"grad_norm": 4.982727527618408,
|
| 780 |
+
"learning_rate": 1.880466472303207e-05,
|
| 781 |
+
"loss": 0.0521,
|
| 782 |
+
"step": 1070
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 3.1486880466472305,
|
| 786 |
+
"grad_norm": 2.4037177562713623,
|
| 787 |
+
"learning_rate": 1.85131195335277e-05,
|
| 788 |
+
"loss": 0.0271,
|
| 789 |
+
"step": 1080
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 3.1778425655976674,
|
| 793 |
+
"grad_norm": 16.846464157104492,
|
| 794 |
+
"learning_rate": 1.8221574344023323e-05,
|
| 795 |
+
"loss": 0.0753,
|
| 796 |
+
"step": 1090
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 3.206997084548105,
|
| 800 |
+
"grad_norm": 0.7762933373451233,
|
| 801 |
+
"learning_rate": 1.793002915451895e-05,
|
| 802 |
+
"loss": 0.0203,
|
| 803 |
+
"step": 1100
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 3.2361516034985423,
|
| 807 |
+
"grad_norm": 0.7675560712814331,
|
| 808 |
+
"learning_rate": 1.7638483965014578e-05,
|
| 809 |
+
"loss": 0.037,
|
| 810 |
+
"step": 1110
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 3.2653061224489797,
|
| 814 |
+
"grad_norm": 6.022889137268066,
|
| 815 |
+
"learning_rate": 1.7346938775510206e-05,
|
| 816 |
+
"loss": 0.0512,
|
| 817 |
+
"step": 1120
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 3.294460641399417,
|
| 821 |
+
"grad_norm": 2.28266978263855,
|
| 822 |
+
"learning_rate": 1.7055393586005833e-05,
|
| 823 |
+
"loss": 0.0594,
|
| 824 |
+
"step": 1130
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 3.323615160349854,
|
| 828 |
+
"grad_norm": 2.531623363494873,
|
| 829 |
+
"learning_rate": 1.6763848396501457e-05,
|
| 830 |
+
"loss": 0.0747,
|
| 831 |
+
"step": 1140
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 3.3527696793002915,
|
| 835 |
+
"grad_norm": 3.3899452686309814,
|
| 836 |
+
"learning_rate": 1.6472303206997085e-05,
|
| 837 |
+
"loss": 0.0716,
|
| 838 |
+
"step": 1150
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 3.381924198250729,
|
| 842 |
+
"grad_norm": 1.235809326171875,
|
| 843 |
+
"learning_rate": 1.6180758017492712e-05,
|
| 844 |
+
"loss": 0.0407,
|
| 845 |
+
"step": 1160
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 3.4110787172011663,
|
| 849 |
+
"grad_norm": 2.9576103687286377,
|
| 850 |
+
"learning_rate": 1.588921282798834e-05,
|
| 851 |
+
"loss": 0.0434,
|
| 852 |
+
"step": 1170
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 3.4402332361516033,
|
| 856 |
+
"grad_norm": 4.579357624053955,
|
| 857 |
+
"learning_rate": 1.5597667638483964e-05,
|
| 858 |
+
"loss": 0.0424,
|
| 859 |
+
"step": 1180
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 3.4693877551020407,
|
| 863 |
+
"grad_norm": 3.1186563968658447,
|
| 864 |
+
"learning_rate": 1.5306122448979594e-05,
|
| 865 |
+
"loss": 0.0269,
|
| 866 |
+
"step": 1190
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 3.498542274052478,
|
| 870 |
+
"grad_norm": 6.906335830688477,
|
| 871 |
+
"learning_rate": 1.5014577259475218e-05,
|
| 872 |
+
"loss": 0.031,
|
| 873 |
+
"step": 1200
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 3.5276967930029155,
|
| 877 |
+
"grad_norm": 1.3179532289505005,
|
| 878 |
+
"learning_rate": 1.4723032069970846e-05,
|
| 879 |
+
"loss": 0.0356,
|
| 880 |
+
"step": 1210
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 3.556851311953353,
|
| 884 |
+
"grad_norm": 2.701486825942993,
|
| 885 |
+
"learning_rate": 1.4431486880466475e-05,
|
| 886 |
+
"loss": 0.0472,
|
| 887 |
+
"step": 1220
|
| 888 |
+
},
|
| 889 |
+
{
|
| 890 |
+
"epoch": 3.5860058309037903,
|
| 891 |
+
"grad_norm": 0.8598970174789429,
|
| 892 |
+
"learning_rate": 1.41399416909621e-05,
|
| 893 |
+
"loss": 0.039,
|
| 894 |
+
"step": 1230
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 3.6151603498542273,
|
| 898 |
+
"grad_norm": 3.872262954711914,
|
| 899 |
+
"learning_rate": 1.3848396501457728e-05,
|
| 900 |
+
"loss": 0.0296,
|
| 901 |
+
"step": 1240
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 3.6443148688046647,
|
| 905 |
+
"grad_norm": 1.517674446105957,
|
| 906 |
+
"learning_rate": 1.3556851311953352e-05,
|
| 907 |
+
"loss": 0.0306,
|
| 908 |
+
"step": 1250
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"epoch": 3.673469387755102,
|
| 912 |
+
"grad_norm": 2.6350300312042236,
|
| 913 |
+
"learning_rate": 1.3265306122448982e-05,
|
| 914 |
+
"loss": 0.0475,
|
| 915 |
+
"step": 1260
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 3.702623906705539,
|
| 919 |
+
"grad_norm": 9.471416473388672,
|
| 920 |
+
"learning_rate": 1.2973760932944606e-05,
|
| 921 |
+
"loss": 0.0366,
|
| 922 |
+
"step": 1270
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 3.7317784256559765,
|
| 926 |
+
"grad_norm": 8.837738037109375,
|
| 927 |
+
"learning_rate": 1.2682215743440235e-05,
|
| 928 |
+
"loss": 0.0493,
|
| 929 |
+
"step": 1280
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 3.760932944606414,
|
| 933 |
+
"grad_norm": 3.3727469444274902,
|
| 934 |
+
"learning_rate": 1.239067055393586e-05,
|
| 935 |
+
"loss": 0.0271,
|
| 936 |
+
"step": 1290
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 3.7900874635568513,
|
| 940 |
+
"grad_norm": 2.860428810119629,
|
| 941 |
+
"learning_rate": 1.2099125364431488e-05,
|
| 942 |
+
"loss": 0.0574,
|
| 943 |
+
"step": 1300
|
| 944 |
+
},
|
| 945 |
+
{
|
| 946 |
+
"epoch": 3.8192419825072887,
|
| 947 |
+
"grad_norm": 12.31369400024414,
|
| 948 |
+
"learning_rate": 1.1807580174927114e-05,
|
| 949 |
+
"loss": 0.0567,
|
| 950 |
+
"step": 1310
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 3.848396501457726,
|
| 954 |
+
"grad_norm": 0.33458849787712097,
|
| 955 |
+
"learning_rate": 1.1516034985422741e-05,
|
| 956 |
+
"loss": 0.0418,
|
| 957 |
+
"step": 1320
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 3.877551020408163,
|
| 961 |
+
"grad_norm": 3.6039276123046875,
|
| 962 |
+
"learning_rate": 1.1224489795918369e-05,
|
| 963 |
+
"loss": 0.0258,
|
| 964 |
+
"step": 1330
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"epoch": 3.9067055393586005,
|
| 968 |
+
"grad_norm": 2.461503505706787,
|
| 969 |
+
"learning_rate": 1.0932944606413994e-05,
|
| 970 |
+
"loss": 0.024,
|
| 971 |
+
"step": 1340
|
| 972 |
+
},
|
| 973 |
+
{
|
| 974 |
+
"epoch": 3.935860058309038,
|
| 975 |
+
"grad_norm": 1.2435688972473145,
|
| 976 |
+
"learning_rate": 1.0641399416909622e-05,
|
| 977 |
+
"loss": 0.0503,
|
| 978 |
+
"step": 1350
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 3.9650145772594754,
|
| 982 |
+
"grad_norm": 6.292943477630615,
|
| 983 |
+
"learning_rate": 1.0349854227405248e-05,
|
| 984 |
+
"loss": 0.037,
|
| 985 |
+
"step": 1360
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 3.9941690962099123,
|
| 989 |
+
"grad_norm": 5.214807033538818,
|
| 990 |
+
"learning_rate": 1.0058309037900875e-05,
|
| 991 |
+
"loss": 0.03,
|
| 992 |
+
"step": 1370
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 4.0,
|
| 996 |
+
"eval_loss": 0.7046686410903931,
|
| 997 |
+
"eval_runtime": 338.8017,
|
| 998 |
+
"eval_samples_per_second": 8.073,
|
| 999 |
+
"eval_steps_per_second": 0.254,
|
| 1000 |
+
"step": 1372
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 4.02332361516035,
|
| 1004 |
+
"grad_norm": 0.912588894367218,
|
| 1005 |
+
"learning_rate": 9.7667638483965e-06,
|
| 1006 |
+
"loss": 0.0185,
|
| 1007 |
+
"step": 1380
|
| 1008 |
+
},
|
| 1009 |
+
{
|
| 1010 |
+
"epoch": 4.052478134110787,
|
| 1011 |
+
"grad_norm": 0.4329352378845215,
|
| 1012 |
+
"learning_rate": 9.47521865889213e-06,
|
| 1013 |
+
"loss": 0.0049,
|
| 1014 |
+
"step": 1390
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 4.081632653061225,
|
| 1018 |
+
"grad_norm": 0.5743452906608582,
|
| 1019 |
+
"learning_rate": 9.183673469387756e-06,
|
| 1020 |
+
"loss": 0.0129,
|
| 1021 |
+
"step": 1400
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"epoch": 4.110787172011662,
|
| 1025 |
+
"grad_norm": 0.7096942663192749,
|
| 1026 |
+
"learning_rate": 8.892128279883383e-06,
|
| 1027 |
+
"loss": 0.0108,
|
| 1028 |
+
"step": 1410
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 4.139941690962099,
|
| 1032 |
+
"grad_norm": 2.971156120300293,
|
| 1033 |
+
"learning_rate": 8.600583090379009e-06,
|
| 1034 |
+
"loss": 0.0171,
|
| 1035 |
+
"step": 1420
|
| 1036 |
+
},
|
| 1037 |
+
{
|
| 1038 |
+
"epoch": 4.169096209912537,
|
| 1039 |
+
"grad_norm": 1.4994933605194092,
|
| 1040 |
+
"learning_rate": 8.309037900874636e-06,
|
| 1041 |
+
"loss": 0.0203,
|
| 1042 |
+
"step": 1430
|
| 1043 |
+
},
|
| 1044 |
+
{
|
| 1045 |
+
"epoch": 4.198250728862973,
|
| 1046 |
+
"grad_norm": 2.779597759246826,
|
| 1047 |
+
"learning_rate": 8.017492711370262e-06,
|
| 1048 |
+
"loss": 0.0115,
|
| 1049 |
+
"step": 1440
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 4.227405247813411,
|
| 1053 |
+
"grad_norm": 1.1518133878707886,
|
| 1054 |
+
"learning_rate": 7.72594752186589e-06,
|
| 1055 |
+
"loss": 0.0099,
|
| 1056 |
+
"step": 1450
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 4.256559766763848,
|
| 1060 |
+
"grad_norm": 0.18404270708560944,
|
| 1061 |
+
"learning_rate": 7.434402332361516e-06,
|
| 1062 |
+
"loss": 0.0123,
|
| 1063 |
+
"step": 1460
|
| 1064 |
+
},
|
| 1065 |
+
{
|
| 1066 |
+
"epoch": 4.285714285714286,
|
| 1067 |
+
"grad_norm": 0.3961102366447449,
|
| 1068 |
+
"learning_rate": 7.142857142857143e-06,
|
| 1069 |
+
"loss": 0.0082,
|
| 1070 |
+
"step": 1470
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"epoch": 4.314868804664723,
|
| 1074 |
+
"grad_norm": 0.31849414110183716,
|
| 1075 |
+
"learning_rate": 6.851311953352769e-06,
|
| 1076 |
+
"loss": 0.0062,
|
| 1077 |
+
"step": 1480
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"epoch": 4.34402332361516,
|
| 1081 |
+
"grad_norm": 0.35347217321395874,
|
| 1082 |
+
"learning_rate": 6.559766763848396e-06,
|
| 1083 |
+
"loss": 0.0091,
|
| 1084 |
+
"step": 1490
|
| 1085 |
+
},
|
| 1086 |
+
{
|
| 1087 |
+
"epoch": 4.373177842565598,
|
| 1088 |
+
"grad_norm": 1.1199325323104858,
|
| 1089 |
+
"learning_rate": 6.268221574344024e-06,
|
| 1090 |
+
"loss": 0.0088,
|
| 1091 |
+
"step": 1500
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 4.402332361516035,
|
| 1095 |
+
"grad_norm": 0.714579701423645,
|
| 1096 |
+
"learning_rate": 5.97667638483965e-06,
|
| 1097 |
+
"loss": 0.0106,
|
| 1098 |
+
"step": 1510
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 4.431486880466473,
|
| 1102 |
+
"grad_norm": 0.4836759865283966,
|
| 1103 |
+
"learning_rate": 5.685131195335277e-06,
|
| 1104 |
+
"loss": 0.0062,
|
| 1105 |
+
"step": 1520
|
| 1106 |
+
},
|
| 1107 |
+
{
|
| 1108 |
+
"epoch": 4.460641399416909,
|
| 1109 |
+
"grad_norm": 1.2272969484329224,
|
| 1110 |
+
"learning_rate": 5.393586005830904e-06,
|
| 1111 |
+
"loss": 0.0134,
|
| 1112 |
+
"step": 1530
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 4.489795918367347,
|
| 1116 |
+
"grad_norm": 1.6942965984344482,
|
| 1117 |
+
"learning_rate": 5.102040816326531e-06,
|
| 1118 |
+
"loss": 0.0158,
|
| 1119 |
+
"step": 1540
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 4.518950437317784,
|
| 1123 |
+
"grad_norm": 7.356329917907715,
|
| 1124 |
+
"learning_rate": 4.810495626822157e-06,
|
| 1125 |
+
"loss": 0.014,
|
| 1126 |
+
"step": 1550
|
| 1127 |
+
},
|
| 1128 |
+
{
|
| 1129 |
+
"epoch": 4.548104956268221,
|
| 1130 |
+
"grad_norm": 0.13148389756679535,
|
| 1131 |
+
"learning_rate": 4.518950437317785e-06,
|
| 1132 |
+
"loss": 0.0078,
|
| 1133 |
+
"step": 1560
|
| 1134 |
+
},
|
| 1135 |
+
{
|
| 1136 |
+
"epoch": 4.577259475218659,
|
| 1137 |
+
"grad_norm": 0.39737656712532043,
|
| 1138 |
+
"learning_rate": 4.227405247813411e-06,
|
| 1139 |
+
"loss": 0.0073,
|
| 1140 |
+
"step": 1570
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"epoch": 4.606413994169096,
|
| 1144 |
+
"grad_norm": 4.271286964416504,
|
| 1145 |
+
"learning_rate": 3.935860058309039e-06,
|
| 1146 |
+
"loss": 0.0076,
|
| 1147 |
+
"step": 1580
|
| 1148 |
+
},
|
| 1149 |
+
{
|
| 1150 |
+
"epoch": 4.635568513119534,
|
| 1151 |
+
"grad_norm": 0.39406758546829224,
|
| 1152 |
+
"learning_rate": 3.644314868804665e-06,
|
| 1153 |
+
"loss": 0.0108,
|
| 1154 |
+
"step": 1590
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 4.664723032069971,
|
| 1158 |
+
"grad_norm": 0.44297096133232117,
|
| 1159 |
+
"learning_rate": 3.352769679300292e-06,
|
| 1160 |
+
"loss": 0.0064,
|
| 1161 |
+
"step": 1600
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 4.6938775510204085,
|
| 1165 |
+
"grad_norm": 3.877448558807373,
|
| 1166 |
+
"learning_rate": 3.0612244897959185e-06,
|
| 1167 |
+
"loss": 0.0094,
|
| 1168 |
+
"step": 1610
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 4.723032069970845,
|
| 1172 |
+
"grad_norm": 0.836543619632721,
|
| 1173 |
+
"learning_rate": 2.7696793002915456e-06,
|
| 1174 |
+
"loss": 0.0093,
|
| 1175 |
+
"step": 1620
|
| 1176 |
+
},
|
| 1177 |
+
{
|
| 1178 |
+
"epoch": 4.752186588921282,
|
| 1179 |
+
"grad_norm": 3.5220887660980225,
|
| 1180 |
+
"learning_rate": 2.478134110787172e-06,
|
| 1181 |
+
"loss": 0.0069,
|
| 1182 |
+
"step": 1630
|
| 1183 |
+
},
|
| 1184 |
+
{
|
| 1185 |
+
"epoch": 4.78134110787172,
|
| 1186 |
+
"grad_norm": 0.4554370641708374,
|
| 1187 |
+
"learning_rate": 2.1865889212827988e-06,
|
| 1188 |
+
"loss": 0.0108,
|
| 1189 |
+
"step": 1640
|
| 1190 |
+
},
|
| 1191 |
+
{
|
| 1192 |
+
"epoch": 4.810495626822157,
|
| 1193 |
+
"grad_norm": 1.146256923675537,
|
| 1194 |
+
"learning_rate": 1.8950437317784258e-06,
|
| 1195 |
+
"loss": 0.0063,
|
| 1196 |
+
"step": 1650
|
| 1197 |
+
},
|
| 1198 |
+
{
|
| 1199 |
+
"epoch": 4.839650145772595,
|
| 1200 |
+
"grad_norm": 0.614227294921875,
|
| 1201 |
+
"learning_rate": 1.6034985422740526e-06,
|
| 1202 |
+
"loss": 0.0117,
|
| 1203 |
+
"step": 1660
|
| 1204 |
+
},
|
| 1205 |
+
{
|
| 1206 |
+
"epoch": 4.868804664723032,
|
| 1207 |
+
"grad_norm": 1.8233929872512817,
|
| 1208 |
+
"learning_rate": 1.3119533527696794e-06,
|
| 1209 |
+
"loss": 0.0094,
|
| 1210 |
+
"step": 1670
|
| 1211 |
+
},
|
| 1212 |
+
{
|
| 1213 |
+
"epoch": 4.8979591836734695,
|
| 1214 |
+
"grad_norm": 1.0853904485702515,
|
| 1215 |
+
"learning_rate": 1.020408163265306e-06,
|
| 1216 |
+
"loss": 0.0105,
|
| 1217 |
+
"step": 1680
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 4.927113702623907,
|
| 1221 |
+
"grad_norm": 0.5117106437683105,
|
| 1222 |
+
"learning_rate": 7.28862973760933e-07,
|
| 1223 |
+
"loss": 0.0064,
|
| 1224 |
+
"step": 1690
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 4.956268221574344,
|
| 1228 |
+
"grad_norm": 1.4317443370819092,
|
| 1229 |
+
"learning_rate": 4.373177842565598e-07,
|
| 1230 |
+
"loss": 0.0051,
|
| 1231 |
+
"step": 1700
|
| 1232 |
+
},
|
| 1233 |
+
{
|
| 1234 |
+
"epoch": 4.985422740524781,
|
| 1235 |
+
"grad_norm": 1.4140545129776,
|
| 1236 |
+
"learning_rate": 1.457725947521866e-07,
|
| 1237 |
+
"loss": 0.009,
|
| 1238 |
+
"step": 1710
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 5.0,
|
| 1242 |
+
"eval_loss": 0.6613177061080933,
|
| 1243 |
+
"eval_runtime": 338.7385,
|
| 1244 |
+
"eval_samples_per_second": 8.074,
|
| 1245 |
+
"eval_steps_per_second": 0.254,
|
| 1246 |
+
"step": 1715
|
| 1247 |
+
}
|
| 1248 |
+
],
|
| 1249 |
+
"logging_steps": 10,
|
| 1250 |
+
"max_steps": 1715,
|
| 1251 |
+
"num_input_tokens_seen": 0,
|
| 1252 |
+
"num_train_epochs": 5,
|
| 1253 |
+
"save_steps": 500,
|
| 1254 |
+
"stateful_callbacks": {
|
| 1255 |
+
"TrainerControl": {
|
| 1256 |
+
"args": {
|
| 1257 |
+
"should_epoch_stop": false,
|
| 1258 |
+
"should_evaluate": false,
|
| 1259 |
+
"should_log": false,
|
| 1260 |
+
"should_save": true,
|
| 1261 |
+
"should_training_stop": true
|
| 1262 |
+
},
|
| 1263 |
+
"attributes": {}
|
| 1264 |
+
}
|
| 1265 |
+
},
|
| 1266 |
+
"total_flos": 5.624623496158249e+18,
|
| 1267 |
+
"train_batch_size": 32,
|
| 1268 |
+
"trial_name": null,
|
| 1269 |
+
"trial_params": null
|
| 1270 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e405555b8a5888fab9a92bf624aac0b7ed3325bfaab4271a6d8bdbfb34fc7012
|
| 3 |
+
size 5304
|