thak123 commited on
Commit
ff19902
1 Parent(s): 447c233

Upload 34 files

Browse files
.gitattributes CHANGED
@@ -34,3 +34,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
  twitter-xlmr-clip-finetuned-all-123/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 
 
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
  twitter-xlmr-clip-finetuned-all-123/tokenizer.json filter=lfs diff=lfs merge=lfs -text
37
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
README.md CHANGED
@@ -1,3 +1,118 @@
1
- ---
2
- license: apache-2.0
3
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: cardiffnlp/twitter-xlm-roberta-base-sentiment-multilingual
3
+ tags:
4
+ - generated_from_trainer
5
+ datasets:
6
+ - all
7
+ metrics:
8
+ - precision
9
+ - recall
10
+ - f1
11
+ model-index:
12
+ - name: twitter-xlmr-clip-finetuned-all-123
13
+ results: []
14
+ ---
15
+
16
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
17
+ should probably proofread and complete it, then remove this comment. -->
18
+
19
+ # twitter-xlmr-clip-finetuned-all-123
20
+
21
+ This model is a fine-tuned version of [cardiffnlp/twitter-xlm-roberta-base-sentiment-multilingual](https://huggingface.co/cardiffnlp/twitter-xlm-roberta-base-sentiment-multilingual) on the all dataset.
22
+ It achieves the following results on the evaluation set:
23
+ - Loss: 0.7405
24
+ - Precision: 0.6431
25
+ - Recall: 0.6554
26
+ - F1: 0.6401
27
+
28
+ ## Model description
29
+
30
+ More information needed
31
+
32
+ ## Intended uses & limitations
33
+
34
+ More information needed
35
+
36
+ ## Training and evaluation data
37
+
38
+ More information needed
39
+
40
+ ## Training procedure
41
+
42
+ ### Training hyperparameters
43
+
44
+ The following hyperparameters were used during training:
45
+ - learning_rate: 5e-05
46
+ - train_batch_size: 16
47
+ - eval_batch_size: 16
48
+ - seed: 123
49
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
50
+ - lr_scheduler_type: linear
51
+ - num_epochs: 50.0
52
+ - mixed_precision_training: Native AMP
53
+
54
+ ### Training results
55
+
56
+ | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 |
57
+ |:-------------:|:-----:|:-----:|:---------------:|:---------:|:------:|:------:|
58
+ | 0.6444 | 0.06 | 500 | 0.8771 | 0.6905 | 0.4537 | 0.4197 |
59
+ | 0.5499 | 0.12 | 1000 | 0.8167 | 0.7197 | 0.4260 | 0.4117 |
60
+ | 0.5357 | 0.18 | 1500 | 0.8084 | 0.7263 | 0.4696 | 0.4424 |
61
+ | 0.5175 | 0.24 | 2000 | 0.8704 | 0.6666 | 0.4266 | 0.3717 |
62
+ | 0.5285 | 0.3 | 2500 | 0.9067 | 0.7529 | 0.4565 | 0.4221 |
63
+ | 0.5081 | 0.36 | 3000 | 0.7414 | 0.7655 | 0.6114 | 0.6356 |
64
+ | 0.506 | 0.42 | 3500 | 0.8713 | 0.5830 | 0.6591 | 0.5786 |
65
+ | 0.5049 | 0.48 | 4000 | 0.7514 | 0.5551 | 0.4568 | 0.4464 |
66
+ | 0.4999 | 0.54 | 4500 | 0.7584 | 0.6519 | 0.5502 | 0.5767 |
67
+ | 0.507 | 0.6 | 5000 | 0.8072 | 0.6479 | 0.5626 | 0.5636 |
68
+ | 0.5048 | 0.66 | 5500 | 0.8080 | 0.6260 | 0.5725 | 0.5730 |
69
+ | 0.4907 | 0.72 | 6000 | 0.7966 | 0.6976 | 0.5138 | 0.5224 |
70
+ | 0.493 | 0.78 | 6500 | 0.8193 | 0.7099 | 0.4949 | 0.4922 |
71
+ | 0.4668 | 0.84 | 7000 | 0.7502 | 0.6282 | 0.6942 | 0.6501 |
72
+ | 0.4717 | 0.9 | 7500 | 0.7636 | 0.6372 | 0.5109 | 0.5191 |
73
+ | 0.4774 | 0.96 | 8000 | 0.7652 | 0.7513 | 0.5360 | 0.5587 |
74
+ | 0.4676 | 1.02 | 8500 | 0.8482 | 0.6372 | 0.5918 | 0.5836 |
75
+ | 0.4361 | 1.08 | 9000 | 0.7456 | 0.6687 | 0.5177 | 0.5175 |
76
+ | 0.4536 | 1.14 | 9500 | 0.8449 | 0.7363 | 0.5160 | 0.5156 |
77
+ | 0.4277 | 1.2 | 10000 | 0.8648 | 0.6382 | 0.5247 | 0.5173 |
78
+ | 0.4444 | 1.26 | 10500 | 0.8723 | 0.5871 | 0.6622 | 0.5959 |
79
+ | 0.4269 | 1.32 | 11000 | 0.7856 | 0.6151 | 0.5521 | 0.5526 |
80
+ | 0.4322 | 1.38 | 11500 | 0.7405 | 0.6431 | 0.6554 | 0.6401 |
81
+ | 0.4435 | 1.44 | 12000 | 0.7682 | 0.6568 | 0.5751 | 0.5923 |
82
+ | 0.4429 | 1.5 | 12500 | 0.8824 | 0.5956 | 0.6006 | 0.5545 |
83
+ | 0.4381 | 1.56 | 13000 | 0.7879 | 0.4457 | 0.4727 | 0.4395 |
84
+ | 0.4389 | 1.62 | 13500 | 0.7555 | 0.6260 | 0.6984 | 0.6502 |
85
+ | 0.4529 | 1.68 | 14000 | 0.7981 | 0.6621 | 0.5546 | 0.5663 |
86
+ | 0.4509 | 1.74 | 14500 | 0.7827 | 0.6160 | 0.6321 | 0.6172 |
87
+ | 0.4413 | 1.8 | 15000 | 0.7895 | 0.6381 | 0.6357 | 0.6285 |
88
+ | 0.4198 | 1.86 | 15500 | 0.8345 | 0.5940 | 0.5526 | 0.5602 |
89
+ | 0.4415 | 1.92 | 16000 | 0.8746 | 0.6615 | 0.6612 | 0.6459 |
90
+ | 0.443 | 1.98 | 16500 | 0.8155 | 0.6516 | 0.5265 | 0.5352 |
91
+ | 0.4068 | 2.04 | 17000 | 0.7642 | 0.5838 | 0.6220 | 0.5975 |
92
+ | 0.3905 | 2.1 | 17500 | 0.7929 | 0.6720 | 0.5555 | 0.5740 |
93
+ | 0.3969 | 2.16 | 18000 | 0.8949 | 0.5330 | 0.4771 | 0.4687 |
94
+ | 0.3841 | 2.22 | 18500 | 0.9233 | 0.6028 | 0.5410 | 0.5492 |
95
+ | 0.4031 | 2.28 | 19000 | 0.7720 | 0.6089 | 0.5719 | 0.5776 |
96
+ | 0.3878 | 2.34 | 19500 | 0.9046 | 0.6265 | 0.5358 | 0.5318 |
97
+ | 0.4001 | 2.41 | 20000 | 0.8451 | 0.6960 | 0.5622 | 0.5761 |
98
+ | 0.3997 | 2.47 | 20500 | 0.8964 | 0.6170 | 0.5665 | 0.5541 |
99
+ | 0.3945 | 2.53 | 21000 | 0.8001 | 0.5553 | 0.5180 | 0.5195 |
100
+ | 0.4005 | 2.59 | 21500 | 0.8357 | 0.5519 | 0.5100 | 0.5170 |
101
+ | 0.3907 | 2.65 | 22000 | 0.8017 | 0.5884 | 0.5409 | 0.5552 |
102
+ | 0.3858 | 2.71 | 22500 | 0.8283 | 0.6036 | 0.5792 | 0.5862 |
103
+ | 0.3973 | 2.77 | 23000 | 0.9024 | 0.5770 | 0.5665 | 0.5393 |
104
+ | 0.3969 | 2.83 | 23500 | 0.8341 | 0.5642 | 0.5528 | 0.5558 |
105
+ | 0.3911 | 2.89 | 24000 | 0.8966 | 0.6045 | 0.5088 | 0.5070 |
106
+ | 0.3856 | 2.95 | 24500 | 0.8349 | 0.6021 | 0.5586 | 0.5689 |
107
+ | 0.3961 | 3.01 | 25000 | 0.9364 | 0.6119 | 0.5412 | 0.5585 |
108
+ | 0.3301 | 3.07 | 25500 | 0.9542 | 0.5757 | 0.6084 | 0.5813 |
109
+ | 0.3385 | 3.13 | 26000 | 1.0137 | 0.5563 | 0.5294 | 0.5346 |
110
+ | 0.3475 | 3.19 | 26500 | 0.9311 | 0.6359 | 0.5675 | 0.5822 |
111
+
112
+
113
+ ### Framework versions
114
+
115
+ - Transformers 4.38.2
116
+ - Pytorch 2.2.1+cu121
117
+ - Datasets 2.18.0
118
+ - Tokenizers 0.15.2
all_results.json ADDED
@@ -0,0 +1,25 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.19,
3
+ "eval_f1": 0.6400986522737416,
4
+ "eval_loss": 0.7404754757881165,
5
+ "eval_precision": 0.6430963017398649,
6
+ "eval_recall": 0.6553870502257598,
7
+ "eval_runtime": 9.726,
8
+ "eval_samples_per_second": 90.273,
9
+ "eval_steps_per_second": 5.655,
10
+ "seed": 123,
11
+ "test_f1": 0.6497165672599908,
12
+ "test_loss": 0.6973698735237122,
13
+ "test_precision": 0.6343420718511593,
14
+ "test_recall": 0.6788425051065489,
15
+ "test_runtime": 68.113,
16
+ "test_samples_per_second": 133.278,
17
+ "test_size": 9078,
18
+ "test_steps_per_second": 8.339,
19
+ "train_loss": 0.44209005161501325,
20
+ "train_runtime": 9420.7211,
21
+ "train_samples_per_second": 706.135,
22
+ "train_size": 133046,
23
+ "train_steps_per_second": 44.137,
24
+ "valid_size": 878
25
+ }
checkpoint-11500/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a63b367b54ba4a9a0c412f0756a7ff6491f8d32109fab990f175ab733091b962
3
+ size 1717381688
checkpoint-11500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b51a238595fe41f46c84af0dd5acf0e0beec929def524f5aa8612b53a31b8724
3
+ size 4197626
checkpoint-11500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5d8c577d58514bee177e5ba2e7161ea18fab4b32823cd3e7898098f29cc248b5
3
+ size 14308
checkpoint-11500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:768a8270714b7317c274ba0a00b243e2d5c9ca8d96fdf90c7e8252c61f836a65
3
+ size 1064
checkpoint-11500/trainer_state.json ADDED
@@ -0,0 +1,435 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.7404754757881165,
3
+ "best_model_checkpoint": "./twitter-xlmr-clip-finetuned-all-123/checkpoint-11500",
4
+ "epoch": 1.3828763828763828,
5
+ "eval_steps": 500,
6
+ "global_step": 11500,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.06,
13
+ "grad_norm": 13.600483894348145,
14
+ "learning_rate": 4.994047619047619e-05,
15
+ "loss": 0.6444,
16
+ "step": 500
17
+ },
18
+ {
19
+ "epoch": 0.06,
20
+ "eval_f1": 0.41968345507991583,
21
+ "eval_loss": 0.8771085143089294,
22
+ "eval_precision": 0.6905442329947281,
23
+ "eval_recall": 0.4537466834241028,
24
+ "eval_runtime": 9.714,
25
+ "eval_samples_per_second": 90.385,
26
+ "eval_steps_per_second": 5.662,
27
+ "step": 500
28
+ },
29
+ {
30
+ "epoch": 0.12,
31
+ "grad_norm": 3.1294376850128174,
32
+ "learning_rate": 4.988035113035113e-05,
33
+ "loss": 0.5499,
34
+ "step": 1000
35
+ },
36
+ {
37
+ "epoch": 0.12,
38
+ "eval_f1": 0.41168390354058104,
39
+ "eval_loss": 0.8166923522949219,
40
+ "eval_precision": 0.7197164083396941,
41
+ "eval_recall": 0.4260270911883815,
42
+ "eval_runtime": 10.112,
43
+ "eval_samples_per_second": 86.827,
44
+ "eval_steps_per_second": 5.439,
45
+ "step": 1000
46
+ },
47
+ {
48
+ "epoch": 0.18,
49
+ "grad_norm": 3.979219913482666,
50
+ "learning_rate": 4.982022607022607e-05,
51
+ "loss": 0.5357,
52
+ "step": 1500
53
+ },
54
+ {
55
+ "epoch": 0.18,
56
+ "eval_f1": 0.44236857569210003,
57
+ "eval_loss": 0.8084450364112854,
58
+ "eval_precision": 0.7263357669536159,
59
+ "eval_recall": 0.4695992179863147,
60
+ "eval_runtime": 9.4795,
61
+ "eval_samples_per_second": 92.621,
62
+ "eval_steps_per_second": 5.802,
63
+ "step": 1500
64
+ },
65
+ {
66
+ "epoch": 0.24,
67
+ "grad_norm": 4.673430442810059,
68
+ "learning_rate": 4.9760341510341516e-05,
69
+ "loss": 0.5175,
70
+ "step": 2000
71
+ },
72
+ {
73
+ "epoch": 0.24,
74
+ "eval_f1": 0.37174117930428513,
75
+ "eval_loss": 0.8703882098197937,
76
+ "eval_precision": 0.6665612592670672,
77
+ "eval_recall": 0.426614532420984,
78
+ "eval_runtime": 10.0742,
79
+ "eval_samples_per_second": 87.153,
80
+ "eval_steps_per_second": 5.459,
81
+ "step": 2000
82
+ },
83
+ {
84
+ "epoch": 0.3,
85
+ "grad_norm": 4.538591384887695,
86
+ "learning_rate": 4.9700216450216456e-05,
87
+ "loss": 0.5285,
88
+ "step": 2500
89
+ },
90
+ {
91
+ "epoch": 0.3,
92
+ "eval_f1": 0.42208412407033724,
93
+ "eval_loss": 0.906663179397583,
94
+ "eval_precision": 0.7528800535330206,
95
+ "eval_recall": 0.45646604291765575,
96
+ "eval_runtime": 11.1647,
97
+ "eval_samples_per_second": 78.641,
98
+ "eval_steps_per_second": 4.926,
99
+ "step": 2500
100
+ },
101
+ {
102
+ "epoch": 0.36,
103
+ "grad_norm": 3.0253360271453857,
104
+ "learning_rate": 4.964009139009139e-05,
105
+ "loss": 0.5081,
106
+ "step": 3000
107
+ },
108
+ {
109
+ "epoch": 0.36,
110
+ "eval_f1": 0.6355885803741356,
111
+ "eval_loss": 0.7414107322692871,
112
+ "eval_precision": 0.7654892081053667,
113
+ "eval_recall": 0.6113652655588139,
114
+ "eval_runtime": 10.1707,
115
+ "eval_samples_per_second": 86.326,
116
+ "eval_steps_per_second": 5.408,
117
+ "step": 3000
118
+ },
119
+ {
120
+ "epoch": 0.42,
121
+ "grad_norm": 7.686459541320801,
122
+ "learning_rate": 4.957996632996633e-05,
123
+ "loss": 0.506,
124
+ "step": 3500
125
+ },
126
+ {
127
+ "epoch": 0.42,
128
+ "eval_f1": 0.5785601517423107,
129
+ "eval_loss": 0.8712885975837708,
130
+ "eval_precision": 0.5829865278619871,
131
+ "eval_recall": 0.659145370758274,
132
+ "eval_runtime": 9.4937,
133
+ "eval_samples_per_second": 92.483,
134
+ "eval_steps_per_second": 5.793,
135
+ "step": 3500
136
+ },
137
+ {
138
+ "epoch": 0.48,
139
+ "grad_norm": 5.288177967071533,
140
+ "learning_rate": 4.951984126984127e-05,
141
+ "loss": 0.5049,
142
+ "step": 4000
143
+ },
144
+ {
145
+ "epoch": 0.48,
146
+ "eval_f1": 0.4464146254881079,
147
+ "eval_loss": 0.751436173915863,
148
+ "eval_precision": 0.5550960735171261,
149
+ "eval_recall": 0.4567546432062561,
150
+ "eval_runtime": 10.1399,
151
+ "eval_samples_per_second": 86.589,
152
+ "eval_steps_per_second": 5.424,
153
+ "step": 4000
154
+ },
155
+ {
156
+ "epoch": 0.54,
157
+ "grad_norm": 2.32623028755188,
158
+ "learning_rate": 4.945971620971621e-05,
159
+ "loss": 0.4999,
160
+ "step": 4500
161
+ },
162
+ {
163
+ "epoch": 0.54,
164
+ "eval_f1": 0.5767450661581325,
165
+ "eval_loss": 0.7584463357925415,
166
+ "eval_precision": 0.6519354622940794,
167
+ "eval_recall": 0.5502052785923753,
168
+ "eval_runtime": 9.964,
169
+ "eval_samples_per_second": 88.117,
170
+ "eval_steps_per_second": 5.52,
171
+ "step": 4500
172
+ },
173
+ {
174
+ "epoch": 0.6,
175
+ "grad_norm": 8.978556632995605,
176
+ "learning_rate": 4.939959114959115e-05,
177
+ "loss": 0.507,
178
+ "step": 5000
179
+ },
180
+ {
181
+ "epoch": 0.6,
182
+ "eval_f1": 0.5635568884134308,
183
+ "eval_loss": 0.8071982860565186,
184
+ "eval_precision": 0.6478587943382395,
185
+ "eval_recall": 0.5625815761299632,
186
+ "eval_runtime": 9.922,
187
+ "eval_samples_per_second": 88.49,
188
+ "eval_steps_per_second": 5.543,
189
+ "step": 5000
190
+ },
191
+ {
192
+ "epoch": 0.66,
193
+ "grad_norm": 5.773237228393555,
194
+ "learning_rate": 4.933946608946609e-05,
195
+ "loss": 0.5048,
196
+ "step": 5500
197
+ },
198
+ {
199
+ "epoch": 0.66,
200
+ "eval_f1": 0.5730277716509274,
201
+ "eval_loss": 0.8080196380615234,
202
+ "eval_precision": 0.6259898141973613,
203
+ "eval_recall": 0.5725429409300378,
204
+ "eval_runtime": 10.1691,
205
+ "eval_samples_per_second": 86.34,
206
+ "eval_steps_per_second": 5.409,
207
+ "step": 5500
208
+ },
209
+ {
210
+ "epoch": 0.72,
211
+ "grad_norm": 4.596500873565674,
212
+ "learning_rate": 4.9279461279461284e-05,
213
+ "loss": 0.4907,
214
+ "step": 6000
215
+ },
216
+ {
217
+ "epoch": 0.72,
218
+ "eval_f1": 0.522356162116029,
219
+ "eval_loss": 0.7966476082801819,
220
+ "eval_precision": 0.6975785731883293,
221
+ "eval_recall": 0.5138146441372248,
222
+ "eval_runtime": 10.1124,
223
+ "eval_samples_per_second": 86.824,
224
+ "eval_steps_per_second": 5.439,
225
+ "step": 6000
226
+ },
227
+ {
228
+ "epoch": 0.78,
229
+ "grad_norm": 4.815864562988281,
230
+ "learning_rate": 4.9219336219336224e-05,
231
+ "loss": 0.493,
232
+ "step": 6500
233
+ },
234
+ {
235
+ "epoch": 0.78,
236
+ "eval_f1": 0.4921730387816156,
237
+ "eval_loss": 0.8192508220672607,
238
+ "eval_precision": 0.7098761086589401,
239
+ "eval_recall": 0.4948889819857562,
240
+ "eval_runtime": 10.12,
241
+ "eval_samples_per_second": 86.759,
242
+ "eval_steps_per_second": 5.435,
243
+ "step": 6500
244
+ },
245
+ {
246
+ "epoch": 0.84,
247
+ "grad_norm": 2.7458622455596924,
248
+ "learning_rate": 4.915921115921116e-05,
249
+ "loss": 0.4668,
250
+ "step": 7000
251
+ },
252
+ {
253
+ "epoch": 0.84,
254
+ "eval_f1": 0.6501053053856183,
255
+ "eval_loss": 0.7501620650291443,
256
+ "eval_precision": 0.6281644378579914,
257
+ "eval_recall": 0.6941954103244425,
258
+ "eval_runtime": 9.7011,
259
+ "eval_samples_per_second": 90.505,
260
+ "eval_steps_per_second": 5.669,
261
+ "step": 7000
262
+ },
263
+ {
264
+ "epoch": 0.9,
265
+ "grad_norm": 6.319346904754639,
266
+ "learning_rate": 4.90990860990861e-05,
267
+ "loss": 0.4717,
268
+ "step": 7500
269
+ },
270
+ {
271
+ "epoch": 0.9,
272
+ "eval_f1": 0.5190561121359437,
273
+ "eval_loss": 0.7636385560035706,
274
+ "eval_precision": 0.6372459240106298,
275
+ "eval_recall": 0.5109072289717451,
276
+ "eval_runtime": 9.9019,
277
+ "eval_samples_per_second": 88.67,
278
+ "eval_steps_per_second": 5.554,
279
+ "step": 7500
280
+ },
281
+ {
282
+ "epoch": 0.96,
283
+ "grad_norm": 4.232970714569092,
284
+ "learning_rate": 4.903896103896104e-05,
285
+ "loss": 0.4774,
286
+ "step": 8000
287
+ },
288
+ {
289
+ "epoch": 0.96,
290
+ "eval_f1": 0.5587125288791309,
291
+ "eval_loss": 0.765232503414154,
292
+ "eval_precision": 0.7513256527294695,
293
+ "eval_recall": 0.5360405902341386,
294
+ "eval_runtime": 10.3268,
295
+ "eval_samples_per_second": 85.022,
296
+ "eval_steps_per_second": 5.326,
297
+ "step": 8000
298
+ },
299
+ {
300
+ "epoch": 1.02,
301
+ "grad_norm": 6.469419479370117,
302
+ "learning_rate": 4.897883597883598e-05,
303
+ "loss": 0.4676,
304
+ "step": 8500
305
+ },
306
+ {
307
+ "epoch": 1.02,
308
+ "eval_f1": 0.5836074412050168,
309
+ "eval_loss": 0.8481860756874084,
310
+ "eval_precision": 0.6371603008082242,
311
+ "eval_recall": 0.5918316808639389,
312
+ "eval_runtime": 10.1025,
313
+ "eval_samples_per_second": 86.909,
314
+ "eval_steps_per_second": 5.444,
315
+ "step": 8500
316
+ },
317
+ {
318
+ "epoch": 1.08,
319
+ "grad_norm": 6.76945161819458,
320
+ "learning_rate": 4.891871091871092e-05,
321
+ "loss": 0.4361,
322
+ "step": 9000
323
+ },
324
+ {
325
+ "epoch": 1.08,
326
+ "eval_f1": 0.5175478304159831,
327
+ "eval_loss": 0.7456216812133789,
328
+ "eval_precision": 0.668652900688299,
329
+ "eval_recall": 0.5176800260671229,
330
+ "eval_runtime": 9.5682,
331
+ "eval_samples_per_second": 91.762,
332
+ "eval_steps_per_second": 5.748,
333
+ "step": 9000
334
+ },
335
+ {
336
+ "epoch": 1.14,
337
+ "grad_norm": 11.62441349029541,
338
+ "learning_rate": 4.885858585858586e-05,
339
+ "loss": 0.4536,
340
+ "step": 9500
341
+ },
342
+ {
343
+ "epoch": 1.14,
344
+ "eval_f1": 0.5155711517804843,
345
+ "eval_loss": 0.8448612689971924,
346
+ "eval_precision": 0.73633642018397,
347
+ "eval_recall": 0.5160256947353722,
348
+ "eval_runtime": 10.0725,
349
+ "eval_samples_per_second": 87.168,
350
+ "eval_steps_per_second": 5.46,
351
+ "step": 9500
352
+ },
353
+ {
354
+ "epoch": 1.2,
355
+ "grad_norm": 6.764497756958008,
356
+ "learning_rate": 4.879858104858105e-05,
357
+ "loss": 0.4277,
358
+ "step": 10000
359
+ },
360
+ {
361
+ "epoch": 1.2,
362
+ "eval_f1": 0.5173349712389866,
363
+ "eval_loss": 0.8647661805152893,
364
+ "eval_precision": 0.6381631231821719,
365
+ "eval_recall": 0.5247414234511009,
366
+ "eval_runtime": 10.0414,
367
+ "eval_samples_per_second": 87.438,
368
+ "eval_steps_per_second": 5.477,
369
+ "step": 10000
370
+ },
371
+ {
372
+ "epoch": 1.26,
373
+ "grad_norm": 6.405815601348877,
374
+ "learning_rate": 4.873845598845599e-05,
375
+ "loss": 0.4444,
376
+ "step": 10500
377
+ },
378
+ {
379
+ "epoch": 1.26,
380
+ "eval_f1": 0.5958880582085585,
381
+ "eval_loss": 0.8722940683364868,
382
+ "eval_precision": 0.5870961573122971,
383
+ "eval_recall": 0.6621933621933621,
384
+ "eval_runtime": 10.0728,
385
+ "eval_samples_per_second": 87.166,
386
+ "eval_steps_per_second": 5.46,
387
+ "step": 10500
388
+ },
389
+ {
390
+ "epoch": 1.32,
391
+ "grad_norm": 3.5875844955444336,
392
+ "learning_rate": 4.8678330928330925e-05,
393
+ "loss": 0.4269,
394
+ "step": 11000
395
+ },
396
+ {
397
+ "epoch": 1.32,
398
+ "eval_f1": 0.5525570018606619,
399
+ "eval_loss": 0.7856002449989319,
400
+ "eval_precision": 0.6150631452676062,
401
+ "eval_recall": 0.5520914211236793,
402
+ "eval_runtime": 9.6968,
403
+ "eval_samples_per_second": 90.545,
404
+ "eval_steps_per_second": 5.672,
405
+ "step": 11000
406
+ },
407
+ {
408
+ "epoch": 1.38,
409
+ "grad_norm": 8.949612617492676,
410
+ "learning_rate": 4.8618205868205866e-05,
411
+ "loss": 0.4322,
412
+ "step": 11500
413
+ },
414
+ {
415
+ "epoch": 1.38,
416
+ "eval_f1": 0.6400986522737416,
417
+ "eval_loss": 0.7404754757881165,
418
+ "eval_precision": 0.6430963017398649,
419
+ "eval_recall": 0.6553870502257598,
420
+ "eval_runtime": 10.249,
421
+ "eval_samples_per_second": 85.667,
422
+ "eval_steps_per_second": 5.366,
423
+ "step": 11500
424
+ }
425
+ ],
426
+ "logging_steps": 500,
427
+ "max_steps": 415800,
428
+ "num_input_tokens_seen": 0,
429
+ "num_train_epochs": 50,
430
+ "save_steps": 500,
431
+ "total_flos": 0.0,
432
+ "train_batch_size": 16,
433
+ "trial_name": null,
434
+ "trial_params": null
435
+ }
checkpoint-11500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:395e85f2911b5fc0d75a2f29579390488ea7028271e65f7acc3187d8dd18ec45
3
+ size 4920
checkpoint-26000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2ddf284532e8186695a85a52b1021cc925048df46cf7d5e977b060e705274ec4
3
+ size 1717381688
checkpoint-26000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:843a8c25daf700404e0ce3846b03aac8b37cf6051e59d249603019e0101c25c5
3
+ size 4197626
checkpoint-26000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4f0fe0816b6c8dfb18d57c814d057beb6521979385e38937766353539d1b3bf2
3
+ size 14308
checkpoint-26000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:848f0c816464d6958cf73f21fbe2e3908c8765b5cb14220ef032e126d0c31863
3
+ size 1064
checkpoint-26000/trainer_state.json ADDED
@@ -0,0 +1,957 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.7404754757881165,
3
+ "best_model_checkpoint": "./twitter-xlmr-clip-finetuned-all-123/checkpoint-11500",
4
+ "epoch": 3.1265031265031267,
5
+ "eval_steps": 500,
6
+ "global_step": 26000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.06,
13
+ "grad_norm": 13.600483894348145,
14
+ "learning_rate": 4.994047619047619e-05,
15
+ "loss": 0.6444,
16
+ "step": 500
17
+ },
18
+ {
19
+ "epoch": 0.06,
20
+ "eval_f1": 0.41968345507991583,
21
+ "eval_loss": 0.8771085143089294,
22
+ "eval_precision": 0.6905442329947281,
23
+ "eval_recall": 0.4537466834241028,
24
+ "eval_runtime": 9.714,
25
+ "eval_samples_per_second": 90.385,
26
+ "eval_steps_per_second": 5.662,
27
+ "step": 500
28
+ },
29
+ {
30
+ "epoch": 0.12,
31
+ "grad_norm": 3.1294376850128174,
32
+ "learning_rate": 4.988035113035113e-05,
33
+ "loss": 0.5499,
34
+ "step": 1000
35
+ },
36
+ {
37
+ "epoch": 0.12,
38
+ "eval_f1": 0.41168390354058104,
39
+ "eval_loss": 0.8166923522949219,
40
+ "eval_precision": 0.7197164083396941,
41
+ "eval_recall": 0.4260270911883815,
42
+ "eval_runtime": 10.112,
43
+ "eval_samples_per_second": 86.827,
44
+ "eval_steps_per_second": 5.439,
45
+ "step": 1000
46
+ },
47
+ {
48
+ "epoch": 0.18,
49
+ "grad_norm": 3.979219913482666,
50
+ "learning_rate": 4.982022607022607e-05,
51
+ "loss": 0.5357,
52
+ "step": 1500
53
+ },
54
+ {
55
+ "epoch": 0.18,
56
+ "eval_f1": 0.44236857569210003,
57
+ "eval_loss": 0.8084450364112854,
58
+ "eval_precision": 0.7263357669536159,
59
+ "eval_recall": 0.4695992179863147,
60
+ "eval_runtime": 9.4795,
61
+ "eval_samples_per_second": 92.621,
62
+ "eval_steps_per_second": 5.802,
63
+ "step": 1500
64
+ },
65
+ {
66
+ "epoch": 0.24,
67
+ "grad_norm": 4.673430442810059,
68
+ "learning_rate": 4.9760341510341516e-05,
69
+ "loss": 0.5175,
70
+ "step": 2000
71
+ },
72
+ {
73
+ "epoch": 0.24,
74
+ "eval_f1": 0.37174117930428513,
75
+ "eval_loss": 0.8703882098197937,
76
+ "eval_precision": 0.6665612592670672,
77
+ "eval_recall": 0.426614532420984,
78
+ "eval_runtime": 10.0742,
79
+ "eval_samples_per_second": 87.153,
80
+ "eval_steps_per_second": 5.459,
81
+ "step": 2000
82
+ },
83
+ {
84
+ "epoch": 0.3,
85
+ "grad_norm": 4.538591384887695,
86
+ "learning_rate": 4.9700216450216456e-05,
87
+ "loss": 0.5285,
88
+ "step": 2500
89
+ },
90
+ {
91
+ "epoch": 0.3,
92
+ "eval_f1": 0.42208412407033724,
93
+ "eval_loss": 0.906663179397583,
94
+ "eval_precision": 0.7528800535330206,
95
+ "eval_recall": 0.45646604291765575,
96
+ "eval_runtime": 11.1647,
97
+ "eval_samples_per_second": 78.641,
98
+ "eval_steps_per_second": 4.926,
99
+ "step": 2500
100
+ },
101
+ {
102
+ "epoch": 0.36,
103
+ "grad_norm": 3.0253360271453857,
104
+ "learning_rate": 4.964009139009139e-05,
105
+ "loss": 0.5081,
106
+ "step": 3000
107
+ },
108
+ {
109
+ "epoch": 0.36,
110
+ "eval_f1": 0.6355885803741356,
111
+ "eval_loss": 0.7414107322692871,
112
+ "eval_precision": 0.7654892081053667,
113
+ "eval_recall": 0.6113652655588139,
114
+ "eval_runtime": 10.1707,
115
+ "eval_samples_per_second": 86.326,
116
+ "eval_steps_per_second": 5.408,
117
+ "step": 3000
118
+ },
119
+ {
120
+ "epoch": 0.42,
121
+ "grad_norm": 7.686459541320801,
122
+ "learning_rate": 4.957996632996633e-05,
123
+ "loss": 0.506,
124
+ "step": 3500
125
+ },
126
+ {
127
+ "epoch": 0.42,
128
+ "eval_f1": 0.5785601517423107,
129
+ "eval_loss": 0.8712885975837708,
130
+ "eval_precision": 0.5829865278619871,
131
+ "eval_recall": 0.659145370758274,
132
+ "eval_runtime": 9.4937,
133
+ "eval_samples_per_second": 92.483,
134
+ "eval_steps_per_second": 5.793,
135
+ "step": 3500
136
+ },
137
+ {
138
+ "epoch": 0.48,
139
+ "grad_norm": 5.288177967071533,
140
+ "learning_rate": 4.951984126984127e-05,
141
+ "loss": 0.5049,
142
+ "step": 4000
143
+ },
144
+ {
145
+ "epoch": 0.48,
146
+ "eval_f1": 0.4464146254881079,
147
+ "eval_loss": 0.751436173915863,
148
+ "eval_precision": 0.5550960735171261,
149
+ "eval_recall": 0.4567546432062561,
150
+ "eval_runtime": 10.1399,
151
+ "eval_samples_per_second": 86.589,
152
+ "eval_steps_per_second": 5.424,
153
+ "step": 4000
154
+ },
155
+ {
156
+ "epoch": 0.54,
157
+ "grad_norm": 2.32623028755188,
158
+ "learning_rate": 4.945971620971621e-05,
159
+ "loss": 0.4999,
160
+ "step": 4500
161
+ },
162
+ {
163
+ "epoch": 0.54,
164
+ "eval_f1": 0.5767450661581325,
165
+ "eval_loss": 0.7584463357925415,
166
+ "eval_precision": 0.6519354622940794,
167
+ "eval_recall": 0.5502052785923753,
168
+ "eval_runtime": 9.964,
169
+ "eval_samples_per_second": 88.117,
170
+ "eval_steps_per_second": 5.52,
171
+ "step": 4500
172
+ },
173
+ {
174
+ "epoch": 0.6,
175
+ "grad_norm": 8.978556632995605,
176
+ "learning_rate": 4.939959114959115e-05,
177
+ "loss": 0.507,
178
+ "step": 5000
179
+ },
180
+ {
181
+ "epoch": 0.6,
182
+ "eval_f1": 0.5635568884134308,
183
+ "eval_loss": 0.8071982860565186,
184
+ "eval_precision": 0.6478587943382395,
185
+ "eval_recall": 0.5625815761299632,
186
+ "eval_runtime": 9.922,
187
+ "eval_samples_per_second": 88.49,
188
+ "eval_steps_per_second": 5.543,
189
+ "step": 5000
190
+ },
191
+ {
192
+ "epoch": 0.66,
193
+ "grad_norm": 5.773237228393555,
194
+ "learning_rate": 4.933946608946609e-05,
195
+ "loss": 0.5048,
196
+ "step": 5500
197
+ },
198
+ {
199
+ "epoch": 0.66,
200
+ "eval_f1": 0.5730277716509274,
201
+ "eval_loss": 0.8080196380615234,
202
+ "eval_precision": 0.6259898141973613,
203
+ "eval_recall": 0.5725429409300378,
204
+ "eval_runtime": 10.1691,
205
+ "eval_samples_per_second": 86.34,
206
+ "eval_steps_per_second": 5.409,
207
+ "step": 5500
208
+ },
209
+ {
210
+ "epoch": 0.72,
211
+ "grad_norm": 4.596500873565674,
212
+ "learning_rate": 4.9279461279461284e-05,
213
+ "loss": 0.4907,
214
+ "step": 6000
215
+ },
216
+ {
217
+ "epoch": 0.72,
218
+ "eval_f1": 0.522356162116029,
219
+ "eval_loss": 0.7966476082801819,
220
+ "eval_precision": 0.6975785731883293,
221
+ "eval_recall": 0.5138146441372248,
222
+ "eval_runtime": 10.1124,
223
+ "eval_samples_per_second": 86.824,
224
+ "eval_steps_per_second": 5.439,
225
+ "step": 6000
226
+ },
227
+ {
228
+ "epoch": 0.78,
229
+ "grad_norm": 4.815864562988281,
230
+ "learning_rate": 4.9219336219336224e-05,
231
+ "loss": 0.493,
232
+ "step": 6500
233
+ },
234
+ {
235
+ "epoch": 0.78,
236
+ "eval_f1": 0.4921730387816156,
237
+ "eval_loss": 0.8192508220672607,
238
+ "eval_precision": 0.7098761086589401,
239
+ "eval_recall": 0.4948889819857562,
240
+ "eval_runtime": 10.12,
241
+ "eval_samples_per_second": 86.759,
242
+ "eval_steps_per_second": 5.435,
243
+ "step": 6500
244
+ },
245
+ {
246
+ "epoch": 0.84,
247
+ "grad_norm": 2.7458622455596924,
248
+ "learning_rate": 4.915921115921116e-05,
249
+ "loss": 0.4668,
250
+ "step": 7000
251
+ },
252
+ {
253
+ "epoch": 0.84,
254
+ "eval_f1": 0.6501053053856183,
255
+ "eval_loss": 0.7501620650291443,
256
+ "eval_precision": 0.6281644378579914,
257
+ "eval_recall": 0.6941954103244425,
258
+ "eval_runtime": 9.7011,
259
+ "eval_samples_per_second": 90.505,
260
+ "eval_steps_per_second": 5.669,
261
+ "step": 7000
262
+ },
263
+ {
264
+ "epoch": 0.9,
265
+ "grad_norm": 6.319346904754639,
266
+ "learning_rate": 4.90990860990861e-05,
267
+ "loss": 0.4717,
268
+ "step": 7500
269
+ },
270
+ {
271
+ "epoch": 0.9,
272
+ "eval_f1": 0.5190561121359437,
273
+ "eval_loss": 0.7636385560035706,
274
+ "eval_precision": 0.6372459240106298,
275
+ "eval_recall": 0.5109072289717451,
276
+ "eval_runtime": 9.9019,
277
+ "eval_samples_per_second": 88.67,
278
+ "eval_steps_per_second": 5.554,
279
+ "step": 7500
280
+ },
281
+ {
282
+ "epoch": 0.96,
283
+ "grad_norm": 4.232970714569092,
284
+ "learning_rate": 4.903896103896104e-05,
285
+ "loss": 0.4774,
286
+ "step": 8000
287
+ },
288
+ {
289
+ "epoch": 0.96,
290
+ "eval_f1": 0.5587125288791309,
291
+ "eval_loss": 0.765232503414154,
292
+ "eval_precision": 0.7513256527294695,
293
+ "eval_recall": 0.5360405902341386,
294
+ "eval_runtime": 10.3268,
295
+ "eval_samples_per_second": 85.022,
296
+ "eval_steps_per_second": 5.326,
297
+ "step": 8000
298
+ },
299
+ {
300
+ "epoch": 1.02,
301
+ "grad_norm": 6.469419479370117,
302
+ "learning_rate": 4.897883597883598e-05,
303
+ "loss": 0.4676,
304
+ "step": 8500
305
+ },
306
+ {
307
+ "epoch": 1.02,
308
+ "eval_f1": 0.5836074412050168,
309
+ "eval_loss": 0.8481860756874084,
310
+ "eval_precision": 0.6371603008082242,
311
+ "eval_recall": 0.5918316808639389,
312
+ "eval_runtime": 10.1025,
313
+ "eval_samples_per_second": 86.909,
314
+ "eval_steps_per_second": 5.444,
315
+ "step": 8500
316
+ },
317
+ {
318
+ "epoch": 1.08,
319
+ "grad_norm": 6.76945161819458,
320
+ "learning_rate": 4.891871091871092e-05,
321
+ "loss": 0.4361,
322
+ "step": 9000
323
+ },
324
+ {
325
+ "epoch": 1.08,
326
+ "eval_f1": 0.5175478304159831,
327
+ "eval_loss": 0.7456216812133789,
328
+ "eval_precision": 0.668652900688299,
329
+ "eval_recall": 0.5176800260671229,
330
+ "eval_runtime": 9.5682,
331
+ "eval_samples_per_second": 91.762,
332
+ "eval_steps_per_second": 5.748,
333
+ "step": 9000
334
+ },
335
+ {
336
+ "epoch": 1.14,
337
+ "grad_norm": 11.62441349029541,
338
+ "learning_rate": 4.885858585858586e-05,
339
+ "loss": 0.4536,
340
+ "step": 9500
341
+ },
342
+ {
343
+ "epoch": 1.14,
344
+ "eval_f1": 0.5155711517804843,
345
+ "eval_loss": 0.8448612689971924,
346
+ "eval_precision": 0.73633642018397,
347
+ "eval_recall": 0.5160256947353722,
348
+ "eval_runtime": 10.0725,
349
+ "eval_samples_per_second": 87.168,
350
+ "eval_steps_per_second": 5.46,
351
+ "step": 9500
352
+ },
353
+ {
354
+ "epoch": 1.2,
355
+ "grad_norm": 6.764497756958008,
356
+ "learning_rate": 4.879858104858105e-05,
357
+ "loss": 0.4277,
358
+ "step": 10000
359
+ },
360
+ {
361
+ "epoch": 1.2,
362
+ "eval_f1": 0.5173349712389866,
363
+ "eval_loss": 0.8647661805152893,
364
+ "eval_precision": 0.6381631231821719,
365
+ "eval_recall": 0.5247414234511009,
366
+ "eval_runtime": 10.0414,
367
+ "eval_samples_per_second": 87.438,
368
+ "eval_steps_per_second": 5.477,
369
+ "step": 10000
370
+ },
371
+ {
372
+ "epoch": 1.26,
373
+ "grad_norm": 6.405815601348877,
374
+ "learning_rate": 4.873845598845599e-05,
375
+ "loss": 0.4444,
376
+ "step": 10500
377
+ },
378
+ {
379
+ "epoch": 1.26,
380
+ "eval_f1": 0.5958880582085585,
381
+ "eval_loss": 0.8722940683364868,
382
+ "eval_precision": 0.5870961573122971,
383
+ "eval_recall": 0.6621933621933621,
384
+ "eval_runtime": 10.0728,
385
+ "eval_samples_per_second": 87.166,
386
+ "eval_steps_per_second": 5.46,
387
+ "step": 10500
388
+ },
389
+ {
390
+ "epoch": 1.32,
391
+ "grad_norm": 3.5875844955444336,
392
+ "learning_rate": 4.8678330928330925e-05,
393
+ "loss": 0.4269,
394
+ "step": 11000
395
+ },
396
+ {
397
+ "epoch": 1.32,
398
+ "eval_f1": 0.5525570018606619,
399
+ "eval_loss": 0.7856002449989319,
400
+ "eval_precision": 0.6150631452676062,
401
+ "eval_recall": 0.5520914211236793,
402
+ "eval_runtime": 9.6968,
403
+ "eval_samples_per_second": 90.545,
404
+ "eval_steps_per_second": 5.672,
405
+ "step": 11000
406
+ },
407
+ {
408
+ "epoch": 1.38,
409
+ "grad_norm": 8.949612617492676,
410
+ "learning_rate": 4.8618205868205866e-05,
411
+ "loss": 0.4322,
412
+ "step": 11500
413
+ },
414
+ {
415
+ "epoch": 1.38,
416
+ "eval_f1": 0.6400986522737416,
417
+ "eval_loss": 0.7404754757881165,
418
+ "eval_precision": 0.6430963017398649,
419
+ "eval_recall": 0.6553870502257598,
420
+ "eval_runtime": 10.249,
421
+ "eval_samples_per_second": 85.667,
422
+ "eval_steps_per_second": 5.366,
423
+ "step": 11500
424
+ },
425
+ {
426
+ "epoch": 1.44,
427
+ "grad_norm": 4.530700206756592,
428
+ "learning_rate": 4.8558080808080806e-05,
429
+ "loss": 0.4435,
430
+ "step": 12000
431
+ },
432
+ {
433
+ "epoch": 1.44,
434
+ "eval_f1": 0.592282401280643,
435
+ "eval_loss": 0.7682257294654846,
436
+ "eval_precision": 0.6568389647238856,
437
+ "eval_recall": 0.5751068286552158,
438
+ "eval_runtime": 9.2229,
439
+ "eval_samples_per_second": 95.197,
440
+ "eval_steps_per_second": 5.963,
441
+ "step": 12000
442
+ },
443
+ {
444
+ "epoch": 1.5,
445
+ "grad_norm": 8.574906349182129,
446
+ "learning_rate": 4.8497955747955746e-05,
447
+ "loss": 0.4429,
448
+ "step": 12500
449
+ },
450
+ {
451
+ "epoch": 1.5,
452
+ "eval_f1": 0.5544893324569983,
453
+ "eval_loss": 0.8823776245117188,
454
+ "eval_precision": 0.595552820840211,
455
+ "eval_recall": 0.6006423683843037,
456
+ "eval_runtime": 9.7179,
457
+ "eval_samples_per_second": 90.349,
458
+ "eval_steps_per_second": 5.66,
459
+ "step": 12500
460
+ },
461
+ {
462
+ "epoch": 1.56,
463
+ "grad_norm": 5.551417827606201,
464
+ "learning_rate": 4.843783068783069e-05,
465
+ "loss": 0.4381,
466
+ "step": 13000
467
+ },
468
+ {
469
+ "epoch": 1.56,
470
+ "eval_f1": 0.4394867511583576,
471
+ "eval_loss": 0.7878994941711426,
472
+ "eval_precision": 0.44570791527313264,
473
+ "eval_recall": 0.4727226178839082,
474
+ "eval_runtime": 9.3935,
475
+ "eval_samples_per_second": 93.469,
476
+ "eval_steps_per_second": 5.855,
477
+ "step": 13000
478
+ },
479
+ {
480
+ "epoch": 1.62,
481
+ "grad_norm": 3.644160032272339,
482
+ "learning_rate": 4.837770562770563e-05,
483
+ "loss": 0.4389,
484
+ "step": 13500
485
+ },
486
+ {
487
+ "epoch": 1.62,
488
+ "eval_f1": 0.6502064155668764,
489
+ "eval_loss": 0.7554711103439331,
490
+ "eval_precision": 0.625995830541285,
491
+ "eval_recall": 0.6983587022296699,
492
+ "eval_runtime": 10.4343,
493
+ "eval_samples_per_second": 84.145,
494
+ "eval_steps_per_second": 5.271,
495
+ "step": 13500
496
+ },
497
+ {
498
+ "epoch": 1.68,
499
+ "grad_norm": 13.295586585998535,
500
+ "learning_rate": 4.8317580567580574e-05,
501
+ "loss": 0.4529,
502
+ "step": 14000
503
+ },
504
+ {
505
+ "epoch": 1.68,
506
+ "eval_f1": 0.5662591977213387,
507
+ "eval_loss": 0.7981444597244263,
508
+ "eval_precision": 0.6620850014647007,
509
+ "eval_recall": 0.5546497230368198,
510
+ "eval_runtime": 10.4044,
511
+ "eval_samples_per_second": 84.388,
512
+ "eval_steps_per_second": 5.286,
513
+ "step": 14000
514
+ },
515
+ {
516
+ "epoch": 1.74,
517
+ "grad_norm": 5.176441192626953,
518
+ "learning_rate": 4.825757575757576e-05,
519
+ "loss": 0.4509,
520
+ "step": 14500
521
+ },
522
+ {
523
+ "epoch": 1.74,
524
+ "eval_f1": 0.6171621860629642,
525
+ "eval_loss": 0.7826597690582275,
526
+ "eval_precision": 0.6159527120477252,
527
+ "eval_recall": 0.6321109714658103,
528
+ "eval_runtime": 10.9264,
529
+ "eval_samples_per_second": 80.356,
530
+ "eval_steps_per_second": 5.034,
531
+ "step": 14500
532
+ },
533
+ {
534
+ "epoch": 1.8,
535
+ "grad_norm": 5.209570407867432,
536
+ "learning_rate": 4.81974506974507e-05,
537
+ "loss": 0.4413,
538
+ "step": 15000
539
+ },
540
+ {
541
+ "epoch": 1.8,
542
+ "eval_f1": 0.6285207961745803,
543
+ "eval_loss": 0.7894985675811768,
544
+ "eval_precision": 0.6381086170916679,
545
+ "eval_recall": 0.6356514453288648,
546
+ "eval_runtime": 10.0334,
547
+ "eval_samples_per_second": 87.508,
548
+ "eval_steps_per_second": 5.482,
549
+ "step": 15000
550
+ },
551
+ {
552
+ "epoch": 1.86,
553
+ "grad_norm": 6.420750141143799,
554
+ "learning_rate": 4.813732563732564e-05,
555
+ "loss": 0.4198,
556
+ "step": 15500
557
+ },
558
+ {
559
+ "epoch": 1.86,
560
+ "eval_f1": 0.5602075791622673,
561
+ "eval_loss": 0.8344631195068359,
562
+ "eval_precision": 0.593987244882881,
563
+ "eval_recall": 0.5526006609877578,
564
+ "eval_runtime": 9.8155,
565
+ "eval_samples_per_second": 89.45,
566
+ "eval_steps_per_second": 5.603,
567
+ "step": 15500
568
+ },
569
+ {
570
+ "epoch": 1.92,
571
+ "grad_norm": 18.093042373657227,
572
+ "learning_rate": 4.807720057720058e-05,
573
+ "loss": 0.4415,
574
+ "step": 16000
575
+ },
576
+ {
577
+ "epoch": 1.92,
578
+ "eval_f1": 0.6458964698548891,
579
+ "eval_loss": 0.8745749592781067,
580
+ "eval_precision": 0.6614986284672897,
581
+ "eval_recall": 0.6612456360843457,
582
+ "eval_runtime": 10.324,
583
+ "eval_samples_per_second": 85.044,
584
+ "eval_steps_per_second": 5.327,
585
+ "step": 16000
586
+ },
587
+ {
588
+ "epoch": 1.98,
589
+ "grad_norm": 3.2640116214752197,
590
+ "learning_rate": 4.801707551707552e-05,
591
+ "loss": 0.443,
592
+ "step": 16500
593
+ },
594
+ {
595
+ "epoch": 1.98,
596
+ "eval_f1": 0.5352104934017916,
597
+ "eval_loss": 0.8155062794685364,
598
+ "eval_precision": 0.6515944452522903,
599
+ "eval_recall": 0.5264720942140296,
600
+ "eval_runtime": 10.2184,
601
+ "eval_samples_per_second": 85.923,
602
+ "eval_steps_per_second": 5.382,
603
+ "step": 16500
604
+ },
605
+ {
606
+ "epoch": 2.04,
607
+ "grad_norm": 6.075052738189697,
608
+ "learning_rate": 4.795695045695046e-05,
609
+ "loss": 0.4068,
610
+ "step": 17000
611
+ },
612
+ {
613
+ "epoch": 2.04,
614
+ "eval_f1": 0.5975450943097163,
615
+ "eval_loss": 0.7641533613204956,
616
+ "eval_precision": 0.5837727558066541,
617
+ "eval_recall": 0.6219876181166503,
618
+ "eval_runtime": 9.7472,
619
+ "eval_samples_per_second": 90.077,
620
+ "eval_steps_per_second": 5.643,
621
+ "step": 17000
622
+ },
623
+ {
624
+ "epoch": 2.1,
625
+ "grad_norm": 2.704364538192749,
626
+ "learning_rate": 4.78968253968254e-05,
627
+ "loss": 0.3905,
628
+ "step": 17500
629
+ },
630
+ {
631
+ "epoch": 2.1,
632
+ "eval_f1": 0.573958551424587,
633
+ "eval_loss": 0.7928948998451233,
634
+ "eval_precision": 0.6719886309250943,
635
+ "eval_recall": 0.5555350742447516,
636
+ "eval_runtime": 9.8882,
637
+ "eval_samples_per_second": 88.792,
638
+ "eval_steps_per_second": 5.562,
639
+ "step": 17500
640
+ },
641
+ {
642
+ "epoch": 2.16,
643
+ "grad_norm": 2.702869176864624,
644
+ "learning_rate": 4.783670033670034e-05,
645
+ "loss": 0.3969,
646
+ "step": 18000
647
+ },
648
+ {
649
+ "epoch": 2.16,
650
+ "eval_f1": 0.46873679220558384,
651
+ "eval_loss": 0.8949081897735596,
652
+ "eval_precision": 0.5330288849723479,
653
+ "eval_recall": 0.4771326164874552,
654
+ "eval_runtime": 9.7502,
655
+ "eval_samples_per_second": 90.05,
656
+ "eval_steps_per_second": 5.641,
657
+ "step": 18000
658
+ },
659
+ {
660
+ "epoch": 2.22,
661
+ "grad_norm": 2.651677370071411,
662
+ "learning_rate": 4.777657527657528e-05,
663
+ "loss": 0.3841,
664
+ "step": 18500
665
+ },
666
+ {
667
+ "epoch": 2.22,
668
+ "eval_f1": 0.5492398227556807,
669
+ "eval_loss": 0.9232780933380127,
670
+ "eval_precision": 0.6027874564459931,
671
+ "eval_recall": 0.5410119629474468,
672
+ "eval_runtime": 9.8226,
673
+ "eval_samples_per_second": 89.386,
674
+ "eval_steps_per_second": 5.599,
675
+ "step": 18500
676
+ },
677
+ {
678
+ "epoch": 2.28,
679
+ "grad_norm": 2.9351656436920166,
680
+ "learning_rate": 4.771645021645022e-05,
681
+ "loss": 0.4031,
682
+ "step": 19000
683
+ },
684
+ {
685
+ "epoch": 2.28,
686
+ "eval_f1": 0.5776004425555427,
687
+ "eval_loss": 0.7720491290092468,
688
+ "eval_precision": 0.6088870316942702,
689
+ "eval_recall": 0.5719396732299958,
690
+ "eval_runtime": 9.6849,
691
+ "eval_samples_per_second": 90.656,
692
+ "eval_steps_per_second": 5.679,
693
+ "step": 19000
694
+ },
695
+ {
696
+ "epoch": 2.34,
697
+ "grad_norm": 10.01524829864502,
698
+ "learning_rate": 4.765632515632516e-05,
699
+ "loss": 0.3878,
700
+ "step": 19500
701
+ },
702
+ {
703
+ "epoch": 2.34,
704
+ "eval_f1": 0.5317940446482866,
705
+ "eval_loss": 0.9046021103858948,
706
+ "eval_precision": 0.6265449365449366,
707
+ "eval_recall": 0.5357957454731649,
708
+ "eval_runtime": 9.4111,
709
+ "eval_samples_per_second": 93.294,
710
+ "eval_steps_per_second": 5.844,
711
+ "step": 19500
712
+ },
713
+ {
714
+ "epoch": 2.41,
715
+ "grad_norm": 1.4061155319213867,
716
+ "learning_rate": 4.75962000962001e-05,
717
+ "loss": 0.4001,
718
+ "step": 20000
719
+ },
720
+ {
721
+ "epoch": 2.41,
722
+ "eval_f1": 0.5760553799542051,
723
+ "eval_loss": 0.8451133370399475,
724
+ "eval_precision": 0.6959572062018271,
725
+ "eval_recall": 0.562243634501699,
726
+ "eval_runtime": 9.8527,
727
+ "eval_samples_per_second": 89.112,
728
+ "eval_steps_per_second": 5.582,
729
+ "step": 20000
730
+ },
731
+ {
732
+ "epoch": 2.47,
733
+ "grad_norm": 2.654297113418579,
734
+ "learning_rate": 4.753607503607504e-05,
735
+ "loss": 0.3997,
736
+ "step": 20500
737
+ },
738
+ {
739
+ "epoch": 2.47,
740
+ "eval_f1": 0.5540787433460292,
741
+ "eval_loss": 0.8963695168495178,
742
+ "eval_precision": 0.6169583136571495,
743
+ "eval_recall": 0.5665354000837871,
744
+ "eval_runtime": 9.8314,
745
+ "eval_samples_per_second": 89.306,
746
+ "eval_steps_per_second": 5.594,
747
+ "step": 20500
748
+ },
749
+ {
750
+ "epoch": 2.53,
751
+ "grad_norm": 2.318394184112549,
752
+ "learning_rate": 4.747607022607023e-05,
753
+ "loss": 0.3945,
754
+ "step": 21000
755
+ },
756
+ {
757
+ "epoch": 2.53,
758
+ "eval_f1": 0.5195041284097078,
759
+ "eval_loss": 0.8000777363777161,
760
+ "eval_precision": 0.5553255357437211,
761
+ "eval_recall": 0.5180002792906019,
762
+ "eval_runtime": 10.1975,
763
+ "eval_samples_per_second": 86.099,
764
+ "eval_steps_per_second": 5.393,
765
+ "step": 21000
766
+ },
767
+ {
768
+ "epoch": 2.59,
769
+ "grad_norm": 6.022315502166748,
770
+ "learning_rate": 4.741594516594517e-05,
771
+ "loss": 0.4005,
772
+ "step": 21500
773
+ },
774
+ {
775
+ "epoch": 2.59,
776
+ "eval_f1": 0.5169696711549415,
777
+ "eval_loss": 0.8356830477714539,
778
+ "eval_precision": 0.5518686564828851,
779
+ "eval_recall": 0.5099939487036261,
780
+ "eval_runtime": 9.9411,
781
+ "eval_samples_per_second": 88.32,
782
+ "eval_steps_per_second": 5.533,
783
+ "step": 21500
784
+ },
785
+ {
786
+ "epoch": 2.65,
787
+ "grad_norm": 3.7972123622894287,
788
+ "learning_rate": 4.735582010582011e-05,
789
+ "loss": 0.3907,
790
+ "step": 22000
791
+ },
792
+ {
793
+ "epoch": 2.65,
794
+ "eval_f1": 0.5552096784695296,
795
+ "eval_loss": 0.8016623258590698,
796
+ "eval_precision": 0.5884397258734974,
797
+ "eval_recall": 0.5408509053670344,
798
+ "eval_runtime": 9.5287,
799
+ "eval_samples_per_second": 92.142,
800
+ "eval_steps_per_second": 5.772,
801
+ "step": 22000
802
+ },
803
+ {
804
+ "epoch": 2.71,
805
+ "grad_norm": 6.792105674743652,
806
+ "learning_rate": 4.729569504569505e-05,
807
+ "loss": 0.3858,
808
+ "step": 22500
809
+ },
810
+ {
811
+ "epoch": 2.71,
812
+ "eval_f1": 0.5861875804431639,
813
+ "eval_loss": 0.8283036351203918,
814
+ "eval_precision": 0.6035754454158798,
815
+ "eval_recall": 0.5791807475678443,
816
+ "eval_runtime": 9.9189,
817
+ "eval_samples_per_second": 88.518,
818
+ "eval_steps_per_second": 5.545,
819
+ "step": 22500
820
+ },
821
+ {
822
+ "epoch": 2.77,
823
+ "grad_norm": 1.6373685598373413,
824
+ "learning_rate": 4.723556998556999e-05,
825
+ "loss": 0.3973,
826
+ "step": 23000
827
+ },
828
+ {
829
+ "epoch": 2.77,
830
+ "eval_f1": 0.5393179472998001,
831
+ "eval_loss": 0.9024349451065063,
832
+ "eval_precision": 0.5769620560542147,
833
+ "eval_recall": 0.5664646464646464,
834
+ "eval_runtime": 9.6142,
835
+ "eval_samples_per_second": 91.324,
836
+ "eval_steps_per_second": 5.721,
837
+ "step": 23000
838
+ },
839
+ {
840
+ "epoch": 2.83,
841
+ "grad_norm": 2.8107876777648926,
842
+ "learning_rate": 4.717544492544493e-05,
843
+ "loss": 0.3969,
844
+ "step": 23500
845
+ },
846
+ {
847
+ "epoch": 2.83,
848
+ "eval_f1": 0.5558204883761618,
849
+ "eval_loss": 0.8340879678726196,
850
+ "eval_precision": 0.564202659920544,
851
+ "eval_recall": 0.5527868547223386,
852
+ "eval_runtime": 9.7568,
853
+ "eval_samples_per_second": 89.988,
854
+ "eval_steps_per_second": 5.637,
855
+ "step": 23500
856
+ },
857
+ {
858
+ "epoch": 2.89,
859
+ "grad_norm": 3.1326379776000977,
860
+ "learning_rate": 4.7115319865319865e-05,
861
+ "loss": 0.3911,
862
+ "step": 24000
863
+ },
864
+ {
865
+ "epoch": 2.89,
866
+ "eval_f1": 0.5070306794950141,
867
+ "eval_loss": 0.8966326117515564,
868
+ "eval_precision": 0.6045140236228393,
869
+ "eval_recall": 0.5087874133035424,
870
+ "eval_runtime": 140.3321,
871
+ "eval_samples_per_second": 6.257,
872
+ "eval_steps_per_second": 0.392,
873
+ "step": 24000
874
+ },
875
+ {
876
+ "epoch": 2.95,
877
+ "grad_norm": 4.015880107879639,
878
+ "learning_rate": 4.7055194805194805e-05,
879
+ "loss": 0.3856,
880
+ "step": 24500
881
+ },
882
+ {
883
+ "epoch": 2.95,
884
+ "eval_f1": 0.5688512961191111,
885
+ "eval_loss": 0.8348873853683472,
886
+ "eval_precision": 0.6021456496959962,
887
+ "eval_recall": 0.5586072708653353,
888
+ "eval_runtime": 77.3964,
889
+ "eval_samples_per_second": 11.344,
890
+ "eval_steps_per_second": 0.711,
891
+ "step": 24500
892
+ },
893
+ {
894
+ "epoch": 3.01,
895
+ "grad_norm": 10.381326675415039,
896
+ "learning_rate": 4.699518999519e-05,
897
+ "loss": 0.3961,
898
+ "step": 25000
899
+ },
900
+ {
901
+ "epoch": 3.01,
902
+ "eval_f1": 0.558529842992594,
903
+ "eval_loss": 0.936424195766449,
904
+ "eval_precision": 0.6119355440337201,
905
+ "eval_recall": 0.5411525392170553,
906
+ "eval_runtime": 9.7898,
907
+ "eval_samples_per_second": 89.685,
908
+ "eval_steps_per_second": 5.618,
909
+ "step": 25000
910
+ },
911
+ {
912
+ "epoch": 3.07,
913
+ "grad_norm": 4.446761608123779,
914
+ "learning_rate": 4.693506493506494e-05,
915
+ "loss": 0.3301,
916
+ "step": 25500
917
+ },
918
+ {
919
+ "epoch": 3.07,
920
+ "eval_f1": 0.5812822447385155,
921
+ "eval_loss": 0.9541723728179932,
922
+ "eval_precision": 0.5757327569674934,
923
+ "eval_recall": 0.6084206116464181,
924
+ "eval_runtime": 9.902,
925
+ "eval_samples_per_second": 88.669,
926
+ "eval_steps_per_second": 5.554,
927
+ "step": 25500
928
+ },
929
+ {
930
+ "epoch": 3.13,
931
+ "grad_norm": 4.501039981842041,
932
+ "learning_rate": 4.687493987493988e-05,
933
+ "loss": 0.3385,
934
+ "step": 26000
935
+ },
936
+ {
937
+ "epoch": 3.13,
938
+ "eval_f1": 0.5345940260961672,
939
+ "eval_loss": 1.0137349367141724,
940
+ "eval_precision": 0.5563381916292068,
941
+ "eval_recall": 0.5293767164734907,
942
+ "eval_runtime": 10.1055,
943
+ "eval_samples_per_second": 86.883,
944
+ "eval_steps_per_second": 5.443,
945
+ "step": 26000
946
+ }
947
+ ],
948
+ "logging_steps": 500,
949
+ "max_steps": 415800,
950
+ "num_input_tokens_seen": 0,
951
+ "num_train_epochs": 50,
952
+ "save_steps": 500,
953
+ "total_flos": 0.0,
954
+ "train_batch_size": 16,
955
+ "trial_name": null,
956
+ "trial_params": null
957
+ }
checkpoint-26000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:395e85f2911b5fc0d75a2f29579390488ea7028271e65f7acc3187d8dd18ec45
3
+ size 4920
checkpoint-26500/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:653b92d4525e139374272800e538514d29ddecb470919a15dd38d6b9341fd3d4
3
+ size 1717381688
checkpoint-26500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ca034047161774b7157647459ab40dc423a54ae60c57dd562d15ac69a048d91c
3
+ size 4197626
checkpoint-26500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a44482721f9beec3a750241eb231efb499e4968fed80a216c35f3dd51a0544ee
3
+ size 14308
checkpoint-26500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:396846fbc866bcc056bfd2df2d043840878aad0005cf85419338e5be5d45ce17
3
+ size 1064
checkpoint-26500/trainer_state.json ADDED
@@ -0,0 +1,975 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.7404754757881165,
3
+ "best_model_checkpoint": "./twitter-xlmr-clip-finetuned-all-123/checkpoint-11500",
4
+ "epoch": 3.1866281866281865,
5
+ "eval_steps": 500,
6
+ "global_step": 26500,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.06,
13
+ "grad_norm": 13.600483894348145,
14
+ "learning_rate": 4.994047619047619e-05,
15
+ "loss": 0.6444,
16
+ "step": 500
17
+ },
18
+ {
19
+ "epoch": 0.06,
20
+ "eval_f1": 0.41968345507991583,
21
+ "eval_loss": 0.8771085143089294,
22
+ "eval_precision": 0.6905442329947281,
23
+ "eval_recall": 0.4537466834241028,
24
+ "eval_runtime": 9.714,
25
+ "eval_samples_per_second": 90.385,
26
+ "eval_steps_per_second": 5.662,
27
+ "step": 500
28
+ },
29
+ {
30
+ "epoch": 0.12,
31
+ "grad_norm": 3.1294376850128174,
32
+ "learning_rate": 4.988035113035113e-05,
33
+ "loss": 0.5499,
34
+ "step": 1000
35
+ },
36
+ {
37
+ "epoch": 0.12,
38
+ "eval_f1": 0.41168390354058104,
39
+ "eval_loss": 0.8166923522949219,
40
+ "eval_precision": 0.7197164083396941,
41
+ "eval_recall": 0.4260270911883815,
42
+ "eval_runtime": 10.112,
43
+ "eval_samples_per_second": 86.827,
44
+ "eval_steps_per_second": 5.439,
45
+ "step": 1000
46
+ },
47
+ {
48
+ "epoch": 0.18,
49
+ "grad_norm": 3.979219913482666,
50
+ "learning_rate": 4.982022607022607e-05,
51
+ "loss": 0.5357,
52
+ "step": 1500
53
+ },
54
+ {
55
+ "epoch": 0.18,
56
+ "eval_f1": 0.44236857569210003,
57
+ "eval_loss": 0.8084450364112854,
58
+ "eval_precision": 0.7263357669536159,
59
+ "eval_recall": 0.4695992179863147,
60
+ "eval_runtime": 9.4795,
61
+ "eval_samples_per_second": 92.621,
62
+ "eval_steps_per_second": 5.802,
63
+ "step": 1500
64
+ },
65
+ {
66
+ "epoch": 0.24,
67
+ "grad_norm": 4.673430442810059,
68
+ "learning_rate": 4.9760341510341516e-05,
69
+ "loss": 0.5175,
70
+ "step": 2000
71
+ },
72
+ {
73
+ "epoch": 0.24,
74
+ "eval_f1": 0.37174117930428513,
75
+ "eval_loss": 0.8703882098197937,
76
+ "eval_precision": 0.6665612592670672,
77
+ "eval_recall": 0.426614532420984,
78
+ "eval_runtime": 10.0742,
79
+ "eval_samples_per_second": 87.153,
80
+ "eval_steps_per_second": 5.459,
81
+ "step": 2000
82
+ },
83
+ {
84
+ "epoch": 0.3,
85
+ "grad_norm": 4.538591384887695,
86
+ "learning_rate": 4.9700216450216456e-05,
87
+ "loss": 0.5285,
88
+ "step": 2500
89
+ },
90
+ {
91
+ "epoch": 0.3,
92
+ "eval_f1": 0.42208412407033724,
93
+ "eval_loss": 0.906663179397583,
94
+ "eval_precision": 0.7528800535330206,
95
+ "eval_recall": 0.45646604291765575,
96
+ "eval_runtime": 11.1647,
97
+ "eval_samples_per_second": 78.641,
98
+ "eval_steps_per_second": 4.926,
99
+ "step": 2500
100
+ },
101
+ {
102
+ "epoch": 0.36,
103
+ "grad_norm": 3.0253360271453857,
104
+ "learning_rate": 4.964009139009139e-05,
105
+ "loss": 0.5081,
106
+ "step": 3000
107
+ },
108
+ {
109
+ "epoch": 0.36,
110
+ "eval_f1": 0.6355885803741356,
111
+ "eval_loss": 0.7414107322692871,
112
+ "eval_precision": 0.7654892081053667,
113
+ "eval_recall": 0.6113652655588139,
114
+ "eval_runtime": 10.1707,
115
+ "eval_samples_per_second": 86.326,
116
+ "eval_steps_per_second": 5.408,
117
+ "step": 3000
118
+ },
119
+ {
120
+ "epoch": 0.42,
121
+ "grad_norm": 7.686459541320801,
122
+ "learning_rate": 4.957996632996633e-05,
123
+ "loss": 0.506,
124
+ "step": 3500
125
+ },
126
+ {
127
+ "epoch": 0.42,
128
+ "eval_f1": 0.5785601517423107,
129
+ "eval_loss": 0.8712885975837708,
130
+ "eval_precision": 0.5829865278619871,
131
+ "eval_recall": 0.659145370758274,
132
+ "eval_runtime": 9.4937,
133
+ "eval_samples_per_second": 92.483,
134
+ "eval_steps_per_second": 5.793,
135
+ "step": 3500
136
+ },
137
+ {
138
+ "epoch": 0.48,
139
+ "grad_norm": 5.288177967071533,
140
+ "learning_rate": 4.951984126984127e-05,
141
+ "loss": 0.5049,
142
+ "step": 4000
143
+ },
144
+ {
145
+ "epoch": 0.48,
146
+ "eval_f1": 0.4464146254881079,
147
+ "eval_loss": 0.751436173915863,
148
+ "eval_precision": 0.5550960735171261,
149
+ "eval_recall": 0.4567546432062561,
150
+ "eval_runtime": 10.1399,
151
+ "eval_samples_per_second": 86.589,
152
+ "eval_steps_per_second": 5.424,
153
+ "step": 4000
154
+ },
155
+ {
156
+ "epoch": 0.54,
157
+ "grad_norm": 2.32623028755188,
158
+ "learning_rate": 4.945971620971621e-05,
159
+ "loss": 0.4999,
160
+ "step": 4500
161
+ },
162
+ {
163
+ "epoch": 0.54,
164
+ "eval_f1": 0.5767450661581325,
165
+ "eval_loss": 0.7584463357925415,
166
+ "eval_precision": 0.6519354622940794,
167
+ "eval_recall": 0.5502052785923753,
168
+ "eval_runtime": 9.964,
169
+ "eval_samples_per_second": 88.117,
170
+ "eval_steps_per_second": 5.52,
171
+ "step": 4500
172
+ },
173
+ {
174
+ "epoch": 0.6,
175
+ "grad_norm": 8.978556632995605,
176
+ "learning_rate": 4.939959114959115e-05,
177
+ "loss": 0.507,
178
+ "step": 5000
179
+ },
180
+ {
181
+ "epoch": 0.6,
182
+ "eval_f1": 0.5635568884134308,
183
+ "eval_loss": 0.8071982860565186,
184
+ "eval_precision": 0.6478587943382395,
185
+ "eval_recall": 0.5625815761299632,
186
+ "eval_runtime": 9.922,
187
+ "eval_samples_per_second": 88.49,
188
+ "eval_steps_per_second": 5.543,
189
+ "step": 5000
190
+ },
191
+ {
192
+ "epoch": 0.66,
193
+ "grad_norm": 5.773237228393555,
194
+ "learning_rate": 4.933946608946609e-05,
195
+ "loss": 0.5048,
196
+ "step": 5500
197
+ },
198
+ {
199
+ "epoch": 0.66,
200
+ "eval_f1": 0.5730277716509274,
201
+ "eval_loss": 0.8080196380615234,
202
+ "eval_precision": 0.6259898141973613,
203
+ "eval_recall": 0.5725429409300378,
204
+ "eval_runtime": 10.1691,
205
+ "eval_samples_per_second": 86.34,
206
+ "eval_steps_per_second": 5.409,
207
+ "step": 5500
208
+ },
209
+ {
210
+ "epoch": 0.72,
211
+ "grad_norm": 4.596500873565674,
212
+ "learning_rate": 4.9279461279461284e-05,
213
+ "loss": 0.4907,
214
+ "step": 6000
215
+ },
216
+ {
217
+ "epoch": 0.72,
218
+ "eval_f1": 0.522356162116029,
219
+ "eval_loss": 0.7966476082801819,
220
+ "eval_precision": 0.6975785731883293,
221
+ "eval_recall": 0.5138146441372248,
222
+ "eval_runtime": 10.1124,
223
+ "eval_samples_per_second": 86.824,
224
+ "eval_steps_per_second": 5.439,
225
+ "step": 6000
226
+ },
227
+ {
228
+ "epoch": 0.78,
229
+ "grad_norm": 4.815864562988281,
230
+ "learning_rate": 4.9219336219336224e-05,
231
+ "loss": 0.493,
232
+ "step": 6500
233
+ },
234
+ {
235
+ "epoch": 0.78,
236
+ "eval_f1": 0.4921730387816156,
237
+ "eval_loss": 0.8192508220672607,
238
+ "eval_precision": 0.7098761086589401,
239
+ "eval_recall": 0.4948889819857562,
240
+ "eval_runtime": 10.12,
241
+ "eval_samples_per_second": 86.759,
242
+ "eval_steps_per_second": 5.435,
243
+ "step": 6500
244
+ },
245
+ {
246
+ "epoch": 0.84,
247
+ "grad_norm": 2.7458622455596924,
248
+ "learning_rate": 4.915921115921116e-05,
249
+ "loss": 0.4668,
250
+ "step": 7000
251
+ },
252
+ {
253
+ "epoch": 0.84,
254
+ "eval_f1": 0.6501053053856183,
255
+ "eval_loss": 0.7501620650291443,
256
+ "eval_precision": 0.6281644378579914,
257
+ "eval_recall": 0.6941954103244425,
258
+ "eval_runtime": 9.7011,
259
+ "eval_samples_per_second": 90.505,
260
+ "eval_steps_per_second": 5.669,
261
+ "step": 7000
262
+ },
263
+ {
264
+ "epoch": 0.9,
265
+ "grad_norm": 6.319346904754639,
266
+ "learning_rate": 4.90990860990861e-05,
267
+ "loss": 0.4717,
268
+ "step": 7500
269
+ },
270
+ {
271
+ "epoch": 0.9,
272
+ "eval_f1": 0.5190561121359437,
273
+ "eval_loss": 0.7636385560035706,
274
+ "eval_precision": 0.6372459240106298,
275
+ "eval_recall": 0.5109072289717451,
276
+ "eval_runtime": 9.9019,
277
+ "eval_samples_per_second": 88.67,
278
+ "eval_steps_per_second": 5.554,
279
+ "step": 7500
280
+ },
281
+ {
282
+ "epoch": 0.96,
283
+ "grad_norm": 4.232970714569092,
284
+ "learning_rate": 4.903896103896104e-05,
285
+ "loss": 0.4774,
286
+ "step": 8000
287
+ },
288
+ {
289
+ "epoch": 0.96,
290
+ "eval_f1": 0.5587125288791309,
291
+ "eval_loss": 0.765232503414154,
292
+ "eval_precision": 0.7513256527294695,
293
+ "eval_recall": 0.5360405902341386,
294
+ "eval_runtime": 10.3268,
295
+ "eval_samples_per_second": 85.022,
296
+ "eval_steps_per_second": 5.326,
297
+ "step": 8000
298
+ },
299
+ {
300
+ "epoch": 1.02,
301
+ "grad_norm": 6.469419479370117,
302
+ "learning_rate": 4.897883597883598e-05,
303
+ "loss": 0.4676,
304
+ "step": 8500
305
+ },
306
+ {
307
+ "epoch": 1.02,
308
+ "eval_f1": 0.5836074412050168,
309
+ "eval_loss": 0.8481860756874084,
310
+ "eval_precision": 0.6371603008082242,
311
+ "eval_recall": 0.5918316808639389,
312
+ "eval_runtime": 10.1025,
313
+ "eval_samples_per_second": 86.909,
314
+ "eval_steps_per_second": 5.444,
315
+ "step": 8500
316
+ },
317
+ {
318
+ "epoch": 1.08,
319
+ "grad_norm": 6.76945161819458,
320
+ "learning_rate": 4.891871091871092e-05,
321
+ "loss": 0.4361,
322
+ "step": 9000
323
+ },
324
+ {
325
+ "epoch": 1.08,
326
+ "eval_f1": 0.5175478304159831,
327
+ "eval_loss": 0.7456216812133789,
328
+ "eval_precision": 0.668652900688299,
329
+ "eval_recall": 0.5176800260671229,
330
+ "eval_runtime": 9.5682,
331
+ "eval_samples_per_second": 91.762,
332
+ "eval_steps_per_second": 5.748,
333
+ "step": 9000
334
+ },
335
+ {
336
+ "epoch": 1.14,
337
+ "grad_norm": 11.62441349029541,
338
+ "learning_rate": 4.885858585858586e-05,
339
+ "loss": 0.4536,
340
+ "step": 9500
341
+ },
342
+ {
343
+ "epoch": 1.14,
344
+ "eval_f1": 0.5155711517804843,
345
+ "eval_loss": 0.8448612689971924,
346
+ "eval_precision": 0.73633642018397,
347
+ "eval_recall": 0.5160256947353722,
348
+ "eval_runtime": 10.0725,
349
+ "eval_samples_per_second": 87.168,
350
+ "eval_steps_per_second": 5.46,
351
+ "step": 9500
352
+ },
353
+ {
354
+ "epoch": 1.2,
355
+ "grad_norm": 6.764497756958008,
356
+ "learning_rate": 4.879858104858105e-05,
357
+ "loss": 0.4277,
358
+ "step": 10000
359
+ },
360
+ {
361
+ "epoch": 1.2,
362
+ "eval_f1": 0.5173349712389866,
363
+ "eval_loss": 0.8647661805152893,
364
+ "eval_precision": 0.6381631231821719,
365
+ "eval_recall": 0.5247414234511009,
366
+ "eval_runtime": 10.0414,
367
+ "eval_samples_per_second": 87.438,
368
+ "eval_steps_per_second": 5.477,
369
+ "step": 10000
370
+ },
371
+ {
372
+ "epoch": 1.26,
373
+ "grad_norm": 6.405815601348877,
374
+ "learning_rate": 4.873845598845599e-05,
375
+ "loss": 0.4444,
376
+ "step": 10500
377
+ },
378
+ {
379
+ "epoch": 1.26,
380
+ "eval_f1": 0.5958880582085585,
381
+ "eval_loss": 0.8722940683364868,
382
+ "eval_precision": 0.5870961573122971,
383
+ "eval_recall": 0.6621933621933621,
384
+ "eval_runtime": 10.0728,
385
+ "eval_samples_per_second": 87.166,
386
+ "eval_steps_per_second": 5.46,
387
+ "step": 10500
388
+ },
389
+ {
390
+ "epoch": 1.32,
391
+ "grad_norm": 3.5875844955444336,
392
+ "learning_rate": 4.8678330928330925e-05,
393
+ "loss": 0.4269,
394
+ "step": 11000
395
+ },
396
+ {
397
+ "epoch": 1.32,
398
+ "eval_f1": 0.5525570018606619,
399
+ "eval_loss": 0.7856002449989319,
400
+ "eval_precision": 0.6150631452676062,
401
+ "eval_recall": 0.5520914211236793,
402
+ "eval_runtime": 9.6968,
403
+ "eval_samples_per_second": 90.545,
404
+ "eval_steps_per_second": 5.672,
405
+ "step": 11000
406
+ },
407
+ {
408
+ "epoch": 1.38,
409
+ "grad_norm": 8.949612617492676,
410
+ "learning_rate": 4.8618205868205866e-05,
411
+ "loss": 0.4322,
412
+ "step": 11500
413
+ },
414
+ {
415
+ "epoch": 1.38,
416
+ "eval_f1": 0.6400986522737416,
417
+ "eval_loss": 0.7404754757881165,
418
+ "eval_precision": 0.6430963017398649,
419
+ "eval_recall": 0.6553870502257598,
420
+ "eval_runtime": 10.249,
421
+ "eval_samples_per_second": 85.667,
422
+ "eval_steps_per_second": 5.366,
423
+ "step": 11500
424
+ },
425
+ {
426
+ "epoch": 1.44,
427
+ "grad_norm": 4.530700206756592,
428
+ "learning_rate": 4.8558080808080806e-05,
429
+ "loss": 0.4435,
430
+ "step": 12000
431
+ },
432
+ {
433
+ "epoch": 1.44,
434
+ "eval_f1": 0.592282401280643,
435
+ "eval_loss": 0.7682257294654846,
436
+ "eval_precision": 0.6568389647238856,
437
+ "eval_recall": 0.5751068286552158,
438
+ "eval_runtime": 9.2229,
439
+ "eval_samples_per_second": 95.197,
440
+ "eval_steps_per_second": 5.963,
441
+ "step": 12000
442
+ },
443
+ {
444
+ "epoch": 1.5,
445
+ "grad_norm": 8.574906349182129,
446
+ "learning_rate": 4.8497955747955746e-05,
447
+ "loss": 0.4429,
448
+ "step": 12500
449
+ },
450
+ {
451
+ "epoch": 1.5,
452
+ "eval_f1": 0.5544893324569983,
453
+ "eval_loss": 0.8823776245117188,
454
+ "eval_precision": 0.595552820840211,
455
+ "eval_recall": 0.6006423683843037,
456
+ "eval_runtime": 9.7179,
457
+ "eval_samples_per_second": 90.349,
458
+ "eval_steps_per_second": 5.66,
459
+ "step": 12500
460
+ },
461
+ {
462
+ "epoch": 1.56,
463
+ "grad_norm": 5.551417827606201,
464
+ "learning_rate": 4.843783068783069e-05,
465
+ "loss": 0.4381,
466
+ "step": 13000
467
+ },
468
+ {
469
+ "epoch": 1.56,
470
+ "eval_f1": 0.4394867511583576,
471
+ "eval_loss": 0.7878994941711426,
472
+ "eval_precision": 0.44570791527313264,
473
+ "eval_recall": 0.4727226178839082,
474
+ "eval_runtime": 9.3935,
475
+ "eval_samples_per_second": 93.469,
476
+ "eval_steps_per_second": 5.855,
477
+ "step": 13000
478
+ },
479
+ {
480
+ "epoch": 1.62,
481
+ "grad_norm": 3.644160032272339,
482
+ "learning_rate": 4.837770562770563e-05,
483
+ "loss": 0.4389,
484
+ "step": 13500
485
+ },
486
+ {
487
+ "epoch": 1.62,
488
+ "eval_f1": 0.6502064155668764,
489
+ "eval_loss": 0.7554711103439331,
490
+ "eval_precision": 0.625995830541285,
491
+ "eval_recall": 0.6983587022296699,
492
+ "eval_runtime": 10.4343,
493
+ "eval_samples_per_second": 84.145,
494
+ "eval_steps_per_second": 5.271,
495
+ "step": 13500
496
+ },
497
+ {
498
+ "epoch": 1.68,
499
+ "grad_norm": 13.295586585998535,
500
+ "learning_rate": 4.8317580567580574e-05,
501
+ "loss": 0.4529,
502
+ "step": 14000
503
+ },
504
+ {
505
+ "epoch": 1.68,
506
+ "eval_f1": 0.5662591977213387,
507
+ "eval_loss": 0.7981444597244263,
508
+ "eval_precision": 0.6620850014647007,
509
+ "eval_recall": 0.5546497230368198,
510
+ "eval_runtime": 10.4044,
511
+ "eval_samples_per_second": 84.388,
512
+ "eval_steps_per_second": 5.286,
513
+ "step": 14000
514
+ },
515
+ {
516
+ "epoch": 1.74,
517
+ "grad_norm": 5.176441192626953,
518
+ "learning_rate": 4.825757575757576e-05,
519
+ "loss": 0.4509,
520
+ "step": 14500
521
+ },
522
+ {
523
+ "epoch": 1.74,
524
+ "eval_f1": 0.6171621860629642,
525
+ "eval_loss": 0.7826597690582275,
526
+ "eval_precision": 0.6159527120477252,
527
+ "eval_recall": 0.6321109714658103,
528
+ "eval_runtime": 10.9264,
529
+ "eval_samples_per_second": 80.356,
530
+ "eval_steps_per_second": 5.034,
531
+ "step": 14500
532
+ },
533
+ {
534
+ "epoch": 1.8,
535
+ "grad_norm": 5.209570407867432,
536
+ "learning_rate": 4.81974506974507e-05,
537
+ "loss": 0.4413,
538
+ "step": 15000
539
+ },
540
+ {
541
+ "epoch": 1.8,
542
+ "eval_f1": 0.6285207961745803,
543
+ "eval_loss": 0.7894985675811768,
544
+ "eval_precision": 0.6381086170916679,
545
+ "eval_recall": 0.6356514453288648,
546
+ "eval_runtime": 10.0334,
547
+ "eval_samples_per_second": 87.508,
548
+ "eval_steps_per_second": 5.482,
549
+ "step": 15000
550
+ },
551
+ {
552
+ "epoch": 1.86,
553
+ "grad_norm": 6.420750141143799,
554
+ "learning_rate": 4.813732563732564e-05,
555
+ "loss": 0.4198,
556
+ "step": 15500
557
+ },
558
+ {
559
+ "epoch": 1.86,
560
+ "eval_f1": 0.5602075791622673,
561
+ "eval_loss": 0.8344631195068359,
562
+ "eval_precision": 0.593987244882881,
563
+ "eval_recall": 0.5526006609877578,
564
+ "eval_runtime": 9.8155,
565
+ "eval_samples_per_second": 89.45,
566
+ "eval_steps_per_second": 5.603,
567
+ "step": 15500
568
+ },
569
+ {
570
+ "epoch": 1.92,
571
+ "grad_norm": 18.093042373657227,
572
+ "learning_rate": 4.807720057720058e-05,
573
+ "loss": 0.4415,
574
+ "step": 16000
575
+ },
576
+ {
577
+ "epoch": 1.92,
578
+ "eval_f1": 0.6458964698548891,
579
+ "eval_loss": 0.8745749592781067,
580
+ "eval_precision": 0.6614986284672897,
581
+ "eval_recall": 0.6612456360843457,
582
+ "eval_runtime": 10.324,
583
+ "eval_samples_per_second": 85.044,
584
+ "eval_steps_per_second": 5.327,
585
+ "step": 16000
586
+ },
587
+ {
588
+ "epoch": 1.98,
589
+ "grad_norm": 3.2640116214752197,
590
+ "learning_rate": 4.801707551707552e-05,
591
+ "loss": 0.443,
592
+ "step": 16500
593
+ },
594
+ {
595
+ "epoch": 1.98,
596
+ "eval_f1": 0.5352104934017916,
597
+ "eval_loss": 0.8155062794685364,
598
+ "eval_precision": 0.6515944452522903,
599
+ "eval_recall": 0.5264720942140296,
600
+ "eval_runtime": 10.2184,
601
+ "eval_samples_per_second": 85.923,
602
+ "eval_steps_per_second": 5.382,
603
+ "step": 16500
604
+ },
605
+ {
606
+ "epoch": 2.04,
607
+ "grad_norm": 6.075052738189697,
608
+ "learning_rate": 4.795695045695046e-05,
609
+ "loss": 0.4068,
610
+ "step": 17000
611
+ },
612
+ {
613
+ "epoch": 2.04,
614
+ "eval_f1": 0.5975450943097163,
615
+ "eval_loss": 0.7641533613204956,
616
+ "eval_precision": 0.5837727558066541,
617
+ "eval_recall": 0.6219876181166503,
618
+ "eval_runtime": 9.7472,
619
+ "eval_samples_per_second": 90.077,
620
+ "eval_steps_per_second": 5.643,
621
+ "step": 17000
622
+ },
623
+ {
624
+ "epoch": 2.1,
625
+ "grad_norm": 2.704364538192749,
626
+ "learning_rate": 4.78968253968254e-05,
627
+ "loss": 0.3905,
628
+ "step": 17500
629
+ },
630
+ {
631
+ "epoch": 2.1,
632
+ "eval_f1": 0.573958551424587,
633
+ "eval_loss": 0.7928948998451233,
634
+ "eval_precision": 0.6719886309250943,
635
+ "eval_recall": 0.5555350742447516,
636
+ "eval_runtime": 9.8882,
637
+ "eval_samples_per_second": 88.792,
638
+ "eval_steps_per_second": 5.562,
639
+ "step": 17500
640
+ },
641
+ {
642
+ "epoch": 2.16,
643
+ "grad_norm": 2.702869176864624,
644
+ "learning_rate": 4.783670033670034e-05,
645
+ "loss": 0.3969,
646
+ "step": 18000
647
+ },
648
+ {
649
+ "epoch": 2.16,
650
+ "eval_f1": 0.46873679220558384,
651
+ "eval_loss": 0.8949081897735596,
652
+ "eval_precision": 0.5330288849723479,
653
+ "eval_recall": 0.4771326164874552,
654
+ "eval_runtime": 9.7502,
655
+ "eval_samples_per_second": 90.05,
656
+ "eval_steps_per_second": 5.641,
657
+ "step": 18000
658
+ },
659
+ {
660
+ "epoch": 2.22,
661
+ "grad_norm": 2.651677370071411,
662
+ "learning_rate": 4.777657527657528e-05,
663
+ "loss": 0.3841,
664
+ "step": 18500
665
+ },
666
+ {
667
+ "epoch": 2.22,
668
+ "eval_f1": 0.5492398227556807,
669
+ "eval_loss": 0.9232780933380127,
670
+ "eval_precision": 0.6027874564459931,
671
+ "eval_recall": 0.5410119629474468,
672
+ "eval_runtime": 9.8226,
673
+ "eval_samples_per_second": 89.386,
674
+ "eval_steps_per_second": 5.599,
675
+ "step": 18500
676
+ },
677
+ {
678
+ "epoch": 2.28,
679
+ "grad_norm": 2.9351656436920166,
680
+ "learning_rate": 4.771645021645022e-05,
681
+ "loss": 0.4031,
682
+ "step": 19000
683
+ },
684
+ {
685
+ "epoch": 2.28,
686
+ "eval_f1": 0.5776004425555427,
687
+ "eval_loss": 0.7720491290092468,
688
+ "eval_precision": 0.6088870316942702,
689
+ "eval_recall": 0.5719396732299958,
690
+ "eval_runtime": 9.6849,
691
+ "eval_samples_per_second": 90.656,
692
+ "eval_steps_per_second": 5.679,
693
+ "step": 19000
694
+ },
695
+ {
696
+ "epoch": 2.34,
697
+ "grad_norm": 10.01524829864502,
698
+ "learning_rate": 4.765632515632516e-05,
699
+ "loss": 0.3878,
700
+ "step": 19500
701
+ },
702
+ {
703
+ "epoch": 2.34,
704
+ "eval_f1": 0.5317940446482866,
705
+ "eval_loss": 0.9046021103858948,
706
+ "eval_precision": 0.6265449365449366,
707
+ "eval_recall": 0.5357957454731649,
708
+ "eval_runtime": 9.4111,
709
+ "eval_samples_per_second": 93.294,
710
+ "eval_steps_per_second": 5.844,
711
+ "step": 19500
712
+ },
713
+ {
714
+ "epoch": 2.41,
715
+ "grad_norm": 1.4061155319213867,
716
+ "learning_rate": 4.75962000962001e-05,
717
+ "loss": 0.4001,
718
+ "step": 20000
719
+ },
720
+ {
721
+ "epoch": 2.41,
722
+ "eval_f1": 0.5760553799542051,
723
+ "eval_loss": 0.8451133370399475,
724
+ "eval_precision": 0.6959572062018271,
725
+ "eval_recall": 0.562243634501699,
726
+ "eval_runtime": 9.8527,
727
+ "eval_samples_per_second": 89.112,
728
+ "eval_steps_per_second": 5.582,
729
+ "step": 20000
730
+ },
731
+ {
732
+ "epoch": 2.47,
733
+ "grad_norm": 2.654297113418579,
734
+ "learning_rate": 4.753607503607504e-05,
735
+ "loss": 0.3997,
736
+ "step": 20500
737
+ },
738
+ {
739
+ "epoch": 2.47,
740
+ "eval_f1": 0.5540787433460292,
741
+ "eval_loss": 0.8963695168495178,
742
+ "eval_precision": 0.6169583136571495,
743
+ "eval_recall": 0.5665354000837871,
744
+ "eval_runtime": 9.8314,
745
+ "eval_samples_per_second": 89.306,
746
+ "eval_steps_per_second": 5.594,
747
+ "step": 20500
748
+ },
749
+ {
750
+ "epoch": 2.53,
751
+ "grad_norm": 2.318394184112549,
752
+ "learning_rate": 4.747607022607023e-05,
753
+ "loss": 0.3945,
754
+ "step": 21000
755
+ },
756
+ {
757
+ "epoch": 2.53,
758
+ "eval_f1": 0.5195041284097078,
759
+ "eval_loss": 0.8000777363777161,
760
+ "eval_precision": 0.5553255357437211,
761
+ "eval_recall": 0.5180002792906019,
762
+ "eval_runtime": 10.1975,
763
+ "eval_samples_per_second": 86.099,
764
+ "eval_steps_per_second": 5.393,
765
+ "step": 21000
766
+ },
767
+ {
768
+ "epoch": 2.59,
769
+ "grad_norm": 6.022315502166748,
770
+ "learning_rate": 4.741594516594517e-05,
771
+ "loss": 0.4005,
772
+ "step": 21500
773
+ },
774
+ {
775
+ "epoch": 2.59,
776
+ "eval_f1": 0.5169696711549415,
777
+ "eval_loss": 0.8356830477714539,
778
+ "eval_precision": 0.5518686564828851,
779
+ "eval_recall": 0.5099939487036261,
780
+ "eval_runtime": 9.9411,
781
+ "eval_samples_per_second": 88.32,
782
+ "eval_steps_per_second": 5.533,
783
+ "step": 21500
784
+ },
785
+ {
786
+ "epoch": 2.65,
787
+ "grad_norm": 3.7972123622894287,
788
+ "learning_rate": 4.735582010582011e-05,
789
+ "loss": 0.3907,
790
+ "step": 22000
791
+ },
792
+ {
793
+ "epoch": 2.65,
794
+ "eval_f1": 0.5552096784695296,
795
+ "eval_loss": 0.8016623258590698,
796
+ "eval_precision": 0.5884397258734974,
797
+ "eval_recall": 0.5408509053670344,
798
+ "eval_runtime": 9.5287,
799
+ "eval_samples_per_second": 92.142,
800
+ "eval_steps_per_second": 5.772,
801
+ "step": 22000
802
+ },
803
+ {
804
+ "epoch": 2.71,
805
+ "grad_norm": 6.792105674743652,
806
+ "learning_rate": 4.729569504569505e-05,
807
+ "loss": 0.3858,
808
+ "step": 22500
809
+ },
810
+ {
811
+ "epoch": 2.71,
812
+ "eval_f1": 0.5861875804431639,
813
+ "eval_loss": 0.8283036351203918,
814
+ "eval_precision": 0.6035754454158798,
815
+ "eval_recall": 0.5791807475678443,
816
+ "eval_runtime": 9.9189,
817
+ "eval_samples_per_second": 88.518,
818
+ "eval_steps_per_second": 5.545,
819
+ "step": 22500
820
+ },
821
+ {
822
+ "epoch": 2.77,
823
+ "grad_norm": 1.6373685598373413,
824
+ "learning_rate": 4.723556998556999e-05,
825
+ "loss": 0.3973,
826
+ "step": 23000
827
+ },
828
+ {
829
+ "epoch": 2.77,
830
+ "eval_f1": 0.5393179472998001,
831
+ "eval_loss": 0.9024349451065063,
832
+ "eval_precision": 0.5769620560542147,
833
+ "eval_recall": 0.5664646464646464,
834
+ "eval_runtime": 9.6142,
835
+ "eval_samples_per_second": 91.324,
836
+ "eval_steps_per_second": 5.721,
837
+ "step": 23000
838
+ },
839
+ {
840
+ "epoch": 2.83,
841
+ "grad_norm": 2.8107876777648926,
842
+ "learning_rate": 4.717544492544493e-05,
843
+ "loss": 0.3969,
844
+ "step": 23500
845
+ },
846
+ {
847
+ "epoch": 2.83,
848
+ "eval_f1": 0.5558204883761618,
849
+ "eval_loss": 0.8340879678726196,
850
+ "eval_precision": 0.564202659920544,
851
+ "eval_recall": 0.5527868547223386,
852
+ "eval_runtime": 9.7568,
853
+ "eval_samples_per_second": 89.988,
854
+ "eval_steps_per_second": 5.637,
855
+ "step": 23500
856
+ },
857
+ {
858
+ "epoch": 2.89,
859
+ "grad_norm": 3.1326379776000977,
860
+ "learning_rate": 4.7115319865319865e-05,
861
+ "loss": 0.3911,
862
+ "step": 24000
863
+ },
864
+ {
865
+ "epoch": 2.89,
866
+ "eval_f1": 0.5070306794950141,
867
+ "eval_loss": 0.8966326117515564,
868
+ "eval_precision": 0.6045140236228393,
869
+ "eval_recall": 0.5087874133035424,
870
+ "eval_runtime": 140.3321,
871
+ "eval_samples_per_second": 6.257,
872
+ "eval_steps_per_second": 0.392,
873
+ "step": 24000
874
+ },
875
+ {
876
+ "epoch": 2.95,
877
+ "grad_norm": 4.015880107879639,
878
+ "learning_rate": 4.7055194805194805e-05,
879
+ "loss": 0.3856,
880
+ "step": 24500
881
+ },
882
+ {
883
+ "epoch": 2.95,
884
+ "eval_f1": 0.5688512961191111,
885
+ "eval_loss": 0.8348873853683472,
886
+ "eval_precision": 0.6021456496959962,
887
+ "eval_recall": 0.5586072708653353,
888
+ "eval_runtime": 77.3964,
889
+ "eval_samples_per_second": 11.344,
890
+ "eval_steps_per_second": 0.711,
891
+ "step": 24500
892
+ },
893
+ {
894
+ "epoch": 3.01,
895
+ "grad_norm": 10.381326675415039,
896
+ "learning_rate": 4.699518999519e-05,
897
+ "loss": 0.3961,
898
+ "step": 25000
899
+ },
900
+ {
901
+ "epoch": 3.01,
902
+ "eval_f1": 0.558529842992594,
903
+ "eval_loss": 0.936424195766449,
904
+ "eval_precision": 0.6119355440337201,
905
+ "eval_recall": 0.5411525392170553,
906
+ "eval_runtime": 9.7898,
907
+ "eval_samples_per_second": 89.685,
908
+ "eval_steps_per_second": 5.618,
909
+ "step": 25000
910
+ },
911
+ {
912
+ "epoch": 3.07,
913
+ "grad_norm": 4.446761608123779,
914
+ "learning_rate": 4.693506493506494e-05,
915
+ "loss": 0.3301,
916
+ "step": 25500
917
+ },
918
+ {
919
+ "epoch": 3.07,
920
+ "eval_f1": 0.5812822447385155,
921
+ "eval_loss": 0.9541723728179932,
922
+ "eval_precision": 0.5757327569674934,
923
+ "eval_recall": 0.6084206116464181,
924
+ "eval_runtime": 9.902,
925
+ "eval_samples_per_second": 88.669,
926
+ "eval_steps_per_second": 5.554,
927
+ "step": 25500
928
+ },
929
+ {
930
+ "epoch": 3.13,
931
+ "grad_norm": 4.501039981842041,
932
+ "learning_rate": 4.687493987493988e-05,
933
+ "loss": 0.3385,
934
+ "step": 26000
935
+ },
936
+ {
937
+ "epoch": 3.13,
938
+ "eval_f1": 0.5345940260961672,
939
+ "eval_loss": 1.0137349367141724,
940
+ "eval_precision": 0.5563381916292068,
941
+ "eval_recall": 0.5293767164734907,
942
+ "eval_runtime": 10.1055,
943
+ "eval_samples_per_second": 86.883,
944
+ "eval_steps_per_second": 5.443,
945
+ "step": 26000
946
+ },
947
+ {
948
+ "epoch": 3.19,
949
+ "grad_norm": 1.9400558471679688,
950
+ "learning_rate": 4.681481481481482e-05,
951
+ "loss": 0.3475,
952
+ "step": 26500
953
+ },
954
+ {
955
+ "epoch": 3.19,
956
+ "eval_f1": 0.5822160079727499,
957
+ "eval_loss": 0.9310874938964844,
958
+ "eval_precision": 0.6358738260592424,
959
+ "eval_recall": 0.5675203649397198,
960
+ "eval_runtime": 9.9305,
961
+ "eval_samples_per_second": 88.414,
962
+ "eval_steps_per_second": 5.538,
963
+ "step": 26500
964
+ }
965
+ ],
966
+ "logging_steps": 500,
967
+ "max_steps": 415800,
968
+ "num_input_tokens_seen": 0,
969
+ "num_train_epochs": 50,
970
+ "save_steps": 500,
971
+ "total_flos": 0.0,
972
+ "train_batch_size": 16,
973
+ "trial_name": null,
974
+ "trial_params": null
975
+ }
checkpoint-26500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:395e85f2911b5fc0d75a2f29579390488ea7028271e65f7acc3187d8dd18ec45
3
+ size 4920
eval_results.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.19,
3
+ "eval_f1": 0.6400986522737416,
4
+ "eval_loss": 0.7404754757881165,
5
+ "eval_precision": 0.6430963017398649,
6
+ "eval_recall": 0.6553870502257598,
7
+ "eval_runtime": 9.726,
8
+ "eval_samples_per_second": 90.273,
9
+ "eval_steps_per_second": 5.655,
10
+ "seed": 123,
11
+ "test_size": 9078,
12
+ "train_size": 133046,
13
+ "valid_size": 878
14
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a63b367b54ba4a9a0c412f0756a7ff6491f8d32109fab990f175ab733091b962
3
+ size 1717381688
preprocessor_config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "crop_size": {
3
+ "height": 224,
4
+ "width": 224
5
+ },
6
+ "do_center_crop": true,
7
+ "do_convert_rgb": true,
8
+ "do_normalize": true,
9
+ "do_rescale": true,
10
+ "do_resize": true,
11
+ "image_mean": [
12
+ 0.48145466,
13
+ 0.4578275,
14
+ 0.40821073
15
+ ],
16
+ "image_processor_type": "CLIPImageProcessor",
17
+ "image_std": [
18
+ 0.26862954,
19
+ 0.26130258,
20
+ 0.27577711
21
+ ],
22
+ "resample": 3,
23
+ "rescale_factor": 0.00392156862745098,
24
+ "size": {
25
+ "shortest_edge": 224
26
+ }
27
+ }
runs/Mar08_17-30-20_cleopatra/events.out.tfevents.1709915444.cleopatra ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d91bfcfa223bcb1cd5628140e48904e7d3e9c63e9f97d15672087848221e6b3a
3
+ size 37900
runs/Mar08_17-30-20_cleopatra/events.out.tfevents.1709924877.cleopatra ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3fe943657cef983e1cadec7eddd2471494e9bed05f1e7712b4076913dde27a0f
3
+ size 896
sentencepiece.bpe.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
3
+ size 5069051
special_tokens_map.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "cls_token": {
10
+ "content": "<s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "eos_token": {
17
+ "content": "</s>",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "mask_token": {
24
+ "content": "<mask>",
25
+ "lstrip": true,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "pad_token": {
31
+ "content": "<pad>",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ },
37
+ "sep_token": {
38
+ "content": "</s>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false
43
+ },
44
+ "unk_token": {
45
+ "content": "<unk>",
46
+ "lstrip": false,
47
+ "normalized": false,
48
+ "rstrip": false,
49
+ "single_word": false
50
+ }
51
+ }
test_predictions_with_text.csv ADDED
The diff for this file is too large to render. See raw diff
 
test_results.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.19,
3
+ "test_f1": 0.6497165672599908,
4
+ "test_loss": 0.6973698735237122,
5
+ "test_precision": 0.6343420718511593,
6
+ "test_recall": 0.6788425051065489,
7
+ "test_runtime": 68.113,
8
+ "test_samples_per_second": 133.278,
9
+ "test_steps_per_second": 8.339
10
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d6f76fe13d42f80dcee0cb86a1aeb5f14f8909bb8a8782f7a4a4ad76697ef164
3
+ size 17083021
tokenizer_config.json ADDED
@@ -0,0 +1,61 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<pad>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "250001": {
36
+ "content": "<mask>",
37
+ "lstrip": true,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "bos_token": "<s>",
45
+ "clean_up_tokenization_spaces": true,
46
+ "cls_token": "<s>",
47
+ "eos_token": "</s>",
48
+ "mask_token": "<mask>",
49
+ "max_length": 128,
50
+ "model_max_length": 1000000000000000019884624838656,
51
+ "pad_to_multiple_of": null,
52
+ "pad_token": "<pad>",
53
+ "pad_token_type_id": 0,
54
+ "padding_side": "right",
55
+ "sep_token": "</s>",
56
+ "stride": 0,
57
+ "tokenizer_class": "XLMRobertaTokenizer",
58
+ "truncation_side": "right",
59
+ "truncation_strategy": "longest_first",
60
+ "unk_token": "<unk>"
61
+ }
train_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.19,
3
+ "train_loss": 0.44209005161501325,
4
+ "train_runtime": 9420.7211,
5
+ "train_samples_per_second": 706.135,
6
+ "train_steps_per_second": 44.137
7
+ }
trainer_state.json ADDED
@@ -0,0 +1,984 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.7404754757881165,
3
+ "best_model_checkpoint": "./twitter-xlmr-clip-finetuned-all-123/checkpoint-11500",
4
+ "epoch": 3.1866281866281865,
5
+ "eval_steps": 500,
6
+ "global_step": 26500,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.06,
13
+ "grad_norm": 13.600483894348145,
14
+ "learning_rate": 4.994047619047619e-05,
15
+ "loss": 0.6444,
16
+ "step": 500
17
+ },
18
+ {
19
+ "epoch": 0.06,
20
+ "eval_f1": 0.41968345507991583,
21
+ "eval_loss": 0.8771085143089294,
22
+ "eval_precision": 0.6905442329947281,
23
+ "eval_recall": 0.4537466834241028,
24
+ "eval_runtime": 9.714,
25
+ "eval_samples_per_second": 90.385,
26
+ "eval_steps_per_second": 5.662,
27
+ "step": 500
28
+ },
29
+ {
30
+ "epoch": 0.12,
31
+ "grad_norm": 3.1294376850128174,
32
+ "learning_rate": 4.988035113035113e-05,
33
+ "loss": 0.5499,
34
+ "step": 1000
35
+ },
36
+ {
37
+ "epoch": 0.12,
38
+ "eval_f1": 0.41168390354058104,
39
+ "eval_loss": 0.8166923522949219,
40
+ "eval_precision": 0.7197164083396941,
41
+ "eval_recall": 0.4260270911883815,
42
+ "eval_runtime": 10.112,
43
+ "eval_samples_per_second": 86.827,
44
+ "eval_steps_per_second": 5.439,
45
+ "step": 1000
46
+ },
47
+ {
48
+ "epoch": 0.18,
49
+ "grad_norm": 3.979219913482666,
50
+ "learning_rate": 4.982022607022607e-05,
51
+ "loss": 0.5357,
52
+ "step": 1500
53
+ },
54
+ {
55
+ "epoch": 0.18,
56
+ "eval_f1": 0.44236857569210003,
57
+ "eval_loss": 0.8084450364112854,
58
+ "eval_precision": 0.7263357669536159,
59
+ "eval_recall": 0.4695992179863147,
60
+ "eval_runtime": 9.4795,
61
+ "eval_samples_per_second": 92.621,
62
+ "eval_steps_per_second": 5.802,
63
+ "step": 1500
64
+ },
65
+ {
66
+ "epoch": 0.24,
67
+ "grad_norm": 4.673430442810059,
68
+ "learning_rate": 4.9760341510341516e-05,
69
+ "loss": 0.5175,
70
+ "step": 2000
71
+ },
72
+ {
73
+ "epoch": 0.24,
74
+ "eval_f1": 0.37174117930428513,
75
+ "eval_loss": 0.8703882098197937,
76
+ "eval_precision": 0.6665612592670672,
77
+ "eval_recall": 0.426614532420984,
78
+ "eval_runtime": 10.0742,
79
+ "eval_samples_per_second": 87.153,
80
+ "eval_steps_per_second": 5.459,
81
+ "step": 2000
82
+ },
83
+ {
84
+ "epoch": 0.3,
85
+ "grad_norm": 4.538591384887695,
86
+ "learning_rate": 4.9700216450216456e-05,
87
+ "loss": 0.5285,
88
+ "step": 2500
89
+ },
90
+ {
91
+ "epoch": 0.3,
92
+ "eval_f1": 0.42208412407033724,
93
+ "eval_loss": 0.906663179397583,
94
+ "eval_precision": 0.7528800535330206,
95
+ "eval_recall": 0.45646604291765575,
96
+ "eval_runtime": 11.1647,
97
+ "eval_samples_per_second": 78.641,
98
+ "eval_steps_per_second": 4.926,
99
+ "step": 2500
100
+ },
101
+ {
102
+ "epoch": 0.36,
103
+ "grad_norm": 3.0253360271453857,
104
+ "learning_rate": 4.964009139009139e-05,
105
+ "loss": 0.5081,
106
+ "step": 3000
107
+ },
108
+ {
109
+ "epoch": 0.36,
110
+ "eval_f1": 0.6355885803741356,
111
+ "eval_loss": 0.7414107322692871,
112
+ "eval_precision": 0.7654892081053667,
113
+ "eval_recall": 0.6113652655588139,
114
+ "eval_runtime": 10.1707,
115
+ "eval_samples_per_second": 86.326,
116
+ "eval_steps_per_second": 5.408,
117
+ "step": 3000
118
+ },
119
+ {
120
+ "epoch": 0.42,
121
+ "grad_norm": 7.686459541320801,
122
+ "learning_rate": 4.957996632996633e-05,
123
+ "loss": 0.506,
124
+ "step": 3500
125
+ },
126
+ {
127
+ "epoch": 0.42,
128
+ "eval_f1": 0.5785601517423107,
129
+ "eval_loss": 0.8712885975837708,
130
+ "eval_precision": 0.5829865278619871,
131
+ "eval_recall": 0.659145370758274,
132
+ "eval_runtime": 9.4937,
133
+ "eval_samples_per_second": 92.483,
134
+ "eval_steps_per_second": 5.793,
135
+ "step": 3500
136
+ },
137
+ {
138
+ "epoch": 0.48,
139
+ "grad_norm": 5.288177967071533,
140
+ "learning_rate": 4.951984126984127e-05,
141
+ "loss": 0.5049,
142
+ "step": 4000
143
+ },
144
+ {
145
+ "epoch": 0.48,
146
+ "eval_f1": 0.4464146254881079,
147
+ "eval_loss": 0.751436173915863,
148
+ "eval_precision": 0.5550960735171261,
149
+ "eval_recall": 0.4567546432062561,
150
+ "eval_runtime": 10.1399,
151
+ "eval_samples_per_second": 86.589,
152
+ "eval_steps_per_second": 5.424,
153
+ "step": 4000
154
+ },
155
+ {
156
+ "epoch": 0.54,
157
+ "grad_norm": 2.32623028755188,
158
+ "learning_rate": 4.945971620971621e-05,
159
+ "loss": 0.4999,
160
+ "step": 4500
161
+ },
162
+ {
163
+ "epoch": 0.54,
164
+ "eval_f1": 0.5767450661581325,
165
+ "eval_loss": 0.7584463357925415,
166
+ "eval_precision": 0.6519354622940794,
167
+ "eval_recall": 0.5502052785923753,
168
+ "eval_runtime": 9.964,
169
+ "eval_samples_per_second": 88.117,
170
+ "eval_steps_per_second": 5.52,
171
+ "step": 4500
172
+ },
173
+ {
174
+ "epoch": 0.6,
175
+ "grad_norm": 8.978556632995605,
176
+ "learning_rate": 4.939959114959115e-05,
177
+ "loss": 0.507,
178
+ "step": 5000
179
+ },
180
+ {
181
+ "epoch": 0.6,
182
+ "eval_f1": 0.5635568884134308,
183
+ "eval_loss": 0.8071982860565186,
184
+ "eval_precision": 0.6478587943382395,
185
+ "eval_recall": 0.5625815761299632,
186
+ "eval_runtime": 9.922,
187
+ "eval_samples_per_second": 88.49,
188
+ "eval_steps_per_second": 5.543,
189
+ "step": 5000
190
+ },
191
+ {
192
+ "epoch": 0.66,
193
+ "grad_norm": 5.773237228393555,
194
+ "learning_rate": 4.933946608946609e-05,
195
+ "loss": 0.5048,
196
+ "step": 5500
197
+ },
198
+ {
199
+ "epoch": 0.66,
200
+ "eval_f1": 0.5730277716509274,
201
+ "eval_loss": 0.8080196380615234,
202
+ "eval_precision": 0.6259898141973613,
203
+ "eval_recall": 0.5725429409300378,
204
+ "eval_runtime": 10.1691,
205
+ "eval_samples_per_second": 86.34,
206
+ "eval_steps_per_second": 5.409,
207
+ "step": 5500
208
+ },
209
+ {
210
+ "epoch": 0.72,
211
+ "grad_norm": 4.596500873565674,
212
+ "learning_rate": 4.9279461279461284e-05,
213
+ "loss": 0.4907,
214
+ "step": 6000
215
+ },
216
+ {
217
+ "epoch": 0.72,
218
+ "eval_f1": 0.522356162116029,
219
+ "eval_loss": 0.7966476082801819,
220
+ "eval_precision": 0.6975785731883293,
221
+ "eval_recall": 0.5138146441372248,
222
+ "eval_runtime": 10.1124,
223
+ "eval_samples_per_second": 86.824,
224
+ "eval_steps_per_second": 5.439,
225
+ "step": 6000
226
+ },
227
+ {
228
+ "epoch": 0.78,
229
+ "grad_norm": 4.815864562988281,
230
+ "learning_rate": 4.9219336219336224e-05,
231
+ "loss": 0.493,
232
+ "step": 6500
233
+ },
234
+ {
235
+ "epoch": 0.78,
236
+ "eval_f1": 0.4921730387816156,
237
+ "eval_loss": 0.8192508220672607,
238
+ "eval_precision": 0.7098761086589401,
239
+ "eval_recall": 0.4948889819857562,
240
+ "eval_runtime": 10.12,
241
+ "eval_samples_per_second": 86.759,
242
+ "eval_steps_per_second": 5.435,
243
+ "step": 6500
244
+ },
245
+ {
246
+ "epoch": 0.84,
247
+ "grad_norm": 2.7458622455596924,
248
+ "learning_rate": 4.915921115921116e-05,
249
+ "loss": 0.4668,
250
+ "step": 7000
251
+ },
252
+ {
253
+ "epoch": 0.84,
254
+ "eval_f1": 0.6501053053856183,
255
+ "eval_loss": 0.7501620650291443,
256
+ "eval_precision": 0.6281644378579914,
257
+ "eval_recall": 0.6941954103244425,
258
+ "eval_runtime": 9.7011,
259
+ "eval_samples_per_second": 90.505,
260
+ "eval_steps_per_second": 5.669,
261
+ "step": 7000
262
+ },
263
+ {
264
+ "epoch": 0.9,
265
+ "grad_norm": 6.319346904754639,
266
+ "learning_rate": 4.90990860990861e-05,
267
+ "loss": 0.4717,
268
+ "step": 7500
269
+ },
270
+ {
271
+ "epoch": 0.9,
272
+ "eval_f1": 0.5190561121359437,
273
+ "eval_loss": 0.7636385560035706,
274
+ "eval_precision": 0.6372459240106298,
275
+ "eval_recall": 0.5109072289717451,
276
+ "eval_runtime": 9.9019,
277
+ "eval_samples_per_second": 88.67,
278
+ "eval_steps_per_second": 5.554,
279
+ "step": 7500
280
+ },
281
+ {
282
+ "epoch": 0.96,
283
+ "grad_norm": 4.232970714569092,
284
+ "learning_rate": 4.903896103896104e-05,
285
+ "loss": 0.4774,
286
+ "step": 8000
287
+ },
288
+ {
289
+ "epoch": 0.96,
290
+ "eval_f1": 0.5587125288791309,
291
+ "eval_loss": 0.765232503414154,
292
+ "eval_precision": 0.7513256527294695,
293
+ "eval_recall": 0.5360405902341386,
294
+ "eval_runtime": 10.3268,
295
+ "eval_samples_per_second": 85.022,
296
+ "eval_steps_per_second": 5.326,
297
+ "step": 8000
298
+ },
299
+ {
300
+ "epoch": 1.02,
301
+ "grad_norm": 6.469419479370117,
302
+ "learning_rate": 4.897883597883598e-05,
303
+ "loss": 0.4676,
304
+ "step": 8500
305
+ },
306
+ {
307
+ "epoch": 1.02,
308
+ "eval_f1": 0.5836074412050168,
309
+ "eval_loss": 0.8481860756874084,
310
+ "eval_precision": 0.6371603008082242,
311
+ "eval_recall": 0.5918316808639389,
312
+ "eval_runtime": 10.1025,
313
+ "eval_samples_per_second": 86.909,
314
+ "eval_steps_per_second": 5.444,
315
+ "step": 8500
316
+ },
317
+ {
318
+ "epoch": 1.08,
319
+ "grad_norm": 6.76945161819458,
320
+ "learning_rate": 4.891871091871092e-05,
321
+ "loss": 0.4361,
322
+ "step": 9000
323
+ },
324
+ {
325
+ "epoch": 1.08,
326
+ "eval_f1": 0.5175478304159831,
327
+ "eval_loss": 0.7456216812133789,
328
+ "eval_precision": 0.668652900688299,
329
+ "eval_recall": 0.5176800260671229,
330
+ "eval_runtime": 9.5682,
331
+ "eval_samples_per_second": 91.762,
332
+ "eval_steps_per_second": 5.748,
333
+ "step": 9000
334
+ },
335
+ {
336
+ "epoch": 1.14,
337
+ "grad_norm": 11.62441349029541,
338
+ "learning_rate": 4.885858585858586e-05,
339
+ "loss": 0.4536,
340
+ "step": 9500
341
+ },
342
+ {
343
+ "epoch": 1.14,
344
+ "eval_f1": 0.5155711517804843,
345
+ "eval_loss": 0.8448612689971924,
346
+ "eval_precision": 0.73633642018397,
347
+ "eval_recall": 0.5160256947353722,
348
+ "eval_runtime": 10.0725,
349
+ "eval_samples_per_second": 87.168,
350
+ "eval_steps_per_second": 5.46,
351
+ "step": 9500
352
+ },
353
+ {
354
+ "epoch": 1.2,
355
+ "grad_norm": 6.764497756958008,
356
+ "learning_rate": 4.879858104858105e-05,
357
+ "loss": 0.4277,
358
+ "step": 10000
359
+ },
360
+ {
361
+ "epoch": 1.2,
362
+ "eval_f1": 0.5173349712389866,
363
+ "eval_loss": 0.8647661805152893,
364
+ "eval_precision": 0.6381631231821719,
365
+ "eval_recall": 0.5247414234511009,
366
+ "eval_runtime": 10.0414,
367
+ "eval_samples_per_second": 87.438,
368
+ "eval_steps_per_second": 5.477,
369
+ "step": 10000
370
+ },
371
+ {
372
+ "epoch": 1.26,
373
+ "grad_norm": 6.405815601348877,
374
+ "learning_rate": 4.873845598845599e-05,
375
+ "loss": 0.4444,
376
+ "step": 10500
377
+ },
378
+ {
379
+ "epoch": 1.26,
380
+ "eval_f1": 0.5958880582085585,
381
+ "eval_loss": 0.8722940683364868,
382
+ "eval_precision": 0.5870961573122971,
383
+ "eval_recall": 0.6621933621933621,
384
+ "eval_runtime": 10.0728,
385
+ "eval_samples_per_second": 87.166,
386
+ "eval_steps_per_second": 5.46,
387
+ "step": 10500
388
+ },
389
+ {
390
+ "epoch": 1.32,
391
+ "grad_norm": 3.5875844955444336,
392
+ "learning_rate": 4.8678330928330925e-05,
393
+ "loss": 0.4269,
394
+ "step": 11000
395
+ },
396
+ {
397
+ "epoch": 1.32,
398
+ "eval_f1": 0.5525570018606619,
399
+ "eval_loss": 0.7856002449989319,
400
+ "eval_precision": 0.6150631452676062,
401
+ "eval_recall": 0.5520914211236793,
402
+ "eval_runtime": 9.6968,
403
+ "eval_samples_per_second": 90.545,
404
+ "eval_steps_per_second": 5.672,
405
+ "step": 11000
406
+ },
407
+ {
408
+ "epoch": 1.38,
409
+ "grad_norm": 8.949612617492676,
410
+ "learning_rate": 4.8618205868205866e-05,
411
+ "loss": 0.4322,
412
+ "step": 11500
413
+ },
414
+ {
415
+ "epoch": 1.38,
416
+ "eval_f1": 0.6400986522737416,
417
+ "eval_loss": 0.7404754757881165,
418
+ "eval_precision": 0.6430963017398649,
419
+ "eval_recall": 0.6553870502257598,
420
+ "eval_runtime": 10.249,
421
+ "eval_samples_per_second": 85.667,
422
+ "eval_steps_per_second": 5.366,
423
+ "step": 11500
424
+ },
425
+ {
426
+ "epoch": 1.44,
427
+ "grad_norm": 4.530700206756592,
428
+ "learning_rate": 4.8558080808080806e-05,
429
+ "loss": 0.4435,
430
+ "step": 12000
431
+ },
432
+ {
433
+ "epoch": 1.44,
434
+ "eval_f1": 0.592282401280643,
435
+ "eval_loss": 0.7682257294654846,
436
+ "eval_precision": 0.6568389647238856,
437
+ "eval_recall": 0.5751068286552158,
438
+ "eval_runtime": 9.2229,
439
+ "eval_samples_per_second": 95.197,
440
+ "eval_steps_per_second": 5.963,
441
+ "step": 12000
442
+ },
443
+ {
444
+ "epoch": 1.5,
445
+ "grad_norm": 8.574906349182129,
446
+ "learning_rate": 4.8497955747955746e-05,
447
+ "loss": 0.4429,
448
+ "step": 12500
449
+ },
450
+ {
451
+ "epoch": 1.5,
452
+ "eval_f1": 0.5544893324569983,
453
+ "eval_loss": 0.8823776245117188,
454
+ "eval_precision": 0.595552820840211,
455
+ "eval_recall": 0.6006423683843037,
456
+ "eval_runtime": 9.7179,
457
+ "eval_samples_per_second": 90.349,
458
+ "eval_steps_per_second": 5.66,
459
+ "step": 12500
460
+ },
461
+ {
462
+ "epoch": 1.56,
463
+ "grad_norm": 5.551417827606201,
464
+ "learning_rate": 4.843783068783069e-05,
465
+ "loss": 0.4381,
466
+ "step": 13000
467
+ },
468
+ {
469
+ "epoch": 1.56,
470
+ "eval_f1": 0.4394867511583576,
471
+ "eval_loss": 0.7878994941711426,
472
+ "eval_precision": 0.44570791527313264,
473
+ "eval_recall": 0.4727226178839082,
474
+ "eval_runtime": 9.3935,
475
+ "eval_samples_per_second": 93.469,
476
+ "eval_steps_per_second": 5.855,
477
+ "step": 13000
478
+ },
479
+ {
480
+ "epoch": 1.62,
481
+ "grad_norm": 3.644160032272339,
482
+ "learning_rate": 4.837770562770563e-05,
483
+ "loss": 0.4389,
484
+ "step": 13500
485
+ },
486
+ {
487
+ "epoch": 1.62,
488
+ "eval_f1": 0.6502064155668764,
489
+ "eval_loss": 0.7554711103439331,
490
+ "eval_precision": 0.625995830541285,
491
+ "eval_recall": 0.6983587022296699,
492
+ "eval_runtime": 10.4343,
493
+ "eval_samples_per_second": 84.145,
494
+ "eval_steps_per_second": 5.271,
495
+ "step": 13500
496
+ },
497
+ {
498
+ "epoch": 1.68,
499
+ "grad_norm": 13.295586585998535,
500
+ "learning_rate": 4.8317580567580574e-05,
501
+ "loss": 0.4529,
502
+ "step": 14000
503
+ },
504
+ {
505
+ "epoch": 1.68,
506
+ "eval_f1": 0.5662591977213387,
507
+ "eval_loss": 0.7981444597244263,
508
+ "eval_precision": 0.6620850014647007,
509
+ "eval_recall": 0.5546497230368198,
510
+ "eval_runtime": 10.4044,
511
+ "eval_samples_per_second": 84.388,
512
+ "eval_steps_per_second": 5.286,
513
+ "step": 14000
514
+ },
515
+ {
516
+ "epoch": 1.74,
517
+ "grad_norm": 5.176441192626953,
518
+ "learning_rate": 4.825757575757576e-05,
519
+ "loss": 0.4509,
520
+ "step": 14500
521
+ },
522
+ {
523
+ "epoch": 1.74,
524
+ "eval_f1": 0.6171621860629642,
525
+ "eval_loss": 0.7826597690582275,
526
+ "eval_precision": 0.6159527120477252,
527
+ "eval_recall": 0.6321109714658103,
528
+ "eval_runtime": 10.9264,
529
+ "eval_samples_per_second": 80.356,
530
+ "eval_steps_per_second": 5.034,
531
+ "step": 14500
532
+ },
533
+ {
534
+ "epoch": 1.8,
535
+ "grad_norm": 5.209570407867432,
536
+ "learning_rate": 4.81974506974507e-05,
537
+ "loss": 0.4413,
538
+ "step": 15000
539
+ },
540
+ {
541
+ "epoch": 1.8,
542
+ "eval_f1": 0.6285207961745803,
543
+ "eval_loss": 0.7894985675811768,
544
+ "eval_precision": 0.6381086170916679,
545
+ "eval_recall": 0.6356514453288648,
546
+ "eval_runtime": 10.0334,
547
+ "eval_samples_per_second": 87.508,
548
+ "eval_steps_per_second": 5.482,
549
+ "step": 15000
550
+ },
551
+ {
552
+ "epoch": 1.86,
553
+ "grad_norm": 6.420750141143799,
554
+ "learning_rate": 4.813732563732564e-05,
555
+ "loss": 0.4198,
556
+ "step": 15500
557
+ },
558
+ {
559
+ "epoch": 1.86,
560
+ "eval_f1": 0.5602075791622673,
561
+ "eval_loss": 0.8344631195068359,
562
+ "eval_precision": 0.593987244882881,
563
+ "eval_recall": 0.5526006609877578,
564
+ "eval_runtime": 9.8155,
565
+ "eval_samples_per_second": 89.45,
566
+ "eval_steps_per_second": 5.603,
567
+ "step": 15500
568
+ },
569
+ {
570
+ "epoch": 1.92,
571
+ "grad_norm": 18.093042373657227,
572
+ "learning_rate": 4.807720057720058e-05,
573
+ "loss": 0.4415,
574
+ "step": 16000
575
+ },
576
+ {
577
+ "epoch": 1.92,
578
+ "eval_f1": 0.6458964698548891,
579
+ "eval_loss": 0.8745749592781067,
580
+ "eval_precision": 0.6614986284672897,
581
+ "eval_recall": 0.6612456360843457,
582
+ "eval_runtime": 10.324,
583
+ "eval_samples_per_second": 85.044,
584
+ "eval_steps_per_second": 5.327,
585
+ "step": 16000
586
+ },
587
+ {
588
+ "epoch": 1.98,
589
+ "grad_norm": 3.2640116214752197,
590
+ "learning_rate": 4.801707551707552e-05,
591
+ "loss": 0.443,
592
+ "step": 16500
593
+ },
594
+ {
595
+ "epoch": 1.98,
596
+ "eval_f1": 0.5352104934017916,
597
+ "eval_loss": 0.8155062794685364,
598
+ "eval_precision": 0.6515944452522903,
599
+ "eval_recall": 0.5264720942140296,
600
+ "eval_runtime": 10.2184,
601
+ "eval_samples_per_second": 85.923,
602
+ "eval_steps_per_second": 5.382,
603
+ "step": 16500
604
+ },
605
+ {
606
+ "epoch": 2.04,
607
+ "grad_norm": 6.075052738189697,
608
+ "learning_rate": 4.795695045695046e-05,
609
+ "loss": 0.4068,
610
+ "step": 17000
611
+ },
612
+ {
613
+ "epoch": 2.04,
614
+ "eval_f1": 0.5975450943097163,
615
+ "eval_loss": 0.7641533613204956,
616
+ "eval_precision": 0.5837727558066541,
617
+ "eval_recall": 0.6219876181166503,
618
+ "eval_runtime": 9.7472,
619
+ "eval_samples_per_second": 90.077,
620
+ "eval_steps_per_second": 5.643,
621
+ "step": 17000
622
+ },
623
+ {
624
+ "epoch": 2.1,
625
+ "grad_norm": 2.704364538192749,
626
+ "learning_rate": 4.78968253968254e-05,
627
+ "loss": 0.3905,
628
+ "step": 17500
629
+ },
630
+ {
631
+ "epoch": 2.1,
632
+ "eval_f1": 0.573958551424587,
633
+ "eval_loss": 0.7928948998451233,
634
+ "eval_precision": 0.6719886309250943,
635
+ "eval_recall": 0.5555350742447516,
636
+ "eval_runtime": 9.8882,
637
+ "eval_samples_per_second": 88.792,
638
+ "eval_steps_per_second": 5.562,
639
+ "step": 17500
640
+ },
641
+ {
642
+ "epoch": 2.16,
643
+ "grad_norm": 2.702869176864624,
644
+ "learning_rate": 4.783670033670034e-05,
645
+ "loss": 0.3969,
646
+ "step": 18000
647
+ },
648
+ {
649
+ "epoch": 2.16,
650
+ "eval_f1": 0.46873679220558384,
651
+ "eval_loss": 0.8949081897735596,
652
+ "eval_precision": 0.5330288849723479,
653
+ "eval_recall": 0.4771326164874552,
654
+ "eval_runtime": 9.7502,
655
+ "eval_samples_per_second": 90.05,
656
+ "eval_steps_per_second": 5.641,
657
+ "step": 18000
658
+ },
659
+ {
660
+ "epoch": 2.22,
661
+ "grad_norm": 2.651677370071411,
662
+ "learning_rate": 4.777657527657528e-05,
663
+ "loss": 0.3841,
664
+ "step": 18500
665
+ },
666
+ {
667
+ "epoch": 2.22,
668
+ "eval_f1": 0.5492398227556807,
669
+ "eval_loss": 0.9232780933380127,
670
+ "eval_precision": 0.6027874564459931,
671
+ "eval_recall": 0.5410119629474468,
672
+ "eval_runtime": 9.8226,
673
+ "eval_samples_per_second": 89.386,
674
+ "eval_steps_per_second": 5.599,
675
+ "step": 18500
676
+ },
677
+ {
678
+ "epoch": 2.28,
679
+ "grad_norm": 2.9351656436920166,
680
+ "learning_rate": 4.771645021645022e-05,
681
+ "loss": 0.4031,
682
+ "step": 19000
683
+ },
684
+ {
685
+ "epoch": 2.28,
686
+ "eval_f1": 0.5776004425555427,
687
+ "eval_loss": 0.7720491290092468,
688
+ "eval_precision": 0.6088870316942702,
689
+ "eval_recall": 0.5719396732299958,
690
+ "eval_runtime": 9.6849,
691
+ "eval_samples_per_second": 90.656,
692
+ "eval_steps_per_second": 5.679,
693
+ "step": 19000
694
+ },
695
+ {
696
+ "epoch": 2.34,
697
+ "grad_norm": 10.01524829864502,
698
+ "learning_rate": 4.765632515632516e-05,
699
+ "loss": 0.3878,
700
+ "step": 19500
701
+ },
702
+ {
703
+ "epoch": 2.34,
704
+ "eval_f1": 0.5317940446482866,
705
+ "eval_loss": 0.9046021103858948,
706
+ "eval_precision": 0.6265449365449366,
707
+ "eval_recall": 0.5357957454731649,
708
+ "eval_runtime": 9.4111,
709
+ "eval_samples_per_second": 93.294,
710
+ "eval_steps_per_second": 5.844,
711
+ "step": 19500
712
+ },
713
+ {
714
+ "epoch": 2.41,
715
+ "grad_norm": 1.4061155319213867,
716
+ "learning_rate": 4.75962000962001e-05,
717
+ "loss": 0.4001,
718
+ "step": 20000
719
+ },
720
+ {
721
+ "epoch": 2.41,
722
+ "eval_f1": 0.5760553799542051,
723
+ "eval_loss": 0.8451133370399475,
724
+ "eval_precision": 0.6959572062018271,
725
+ "eval_recall": 0.562243634501699,
726
+ "eval_runtime": 9.8527,
727
+ "eval_samples_per_second": 89.112,
728
+ "eval_steps_per_second": 5.582,
729
+ "step": 20000
730
+ },
731
+ {
732
+ "epoch": 2.47,
733
+ "grad_norm": 2.654297113418579,
734
+ "learning_rate": 4.753607503607504e-05,
735
+ "loss": 0.3997,
736
+ "step": 20500
737
+ },
738
+ {
739
+ "epoch": 2.47,
740
+ "eval_f1": 0.5540787433460292,
741
+ "eval_loss": 0.8963695168495178,
742
+ "eval_precision": 0.6169583136571495,
743
+ "eval_recall": 0.5665354000837871,
744
+ "eval_runtime": 9.8314,
745
+ "eval_samples_per_second": 89.306,
746
+ "eval_steps_per_second": 5.594,
747
+ "step": 20500
748
+ },
749
+ {
750
+ "epoch": 2.53,
751
+ "grad_norm": 2.318394184112549,
752
+ "learning_rate": 4.747607022607023e-05,
753
+ "loss": 0.3945,
754
+ "step": 21000
755
+ },
756
+ {
757
+ "epoch": 2.53,
758
+ "eval_f1": 0.5195041284097078,
759
+ "eval_loss": 0.8000777363777161,
760
+ "eval_precision": 0.5553255357437211,
761
+ "eval_recall": 0.5180002792906019,
762
+ "eval_runtime": 10.1975,
763
+ "eval_samples_per_second": 86.099,
764
+ "eval_steps_per_second": 5.393,
765
+ "step": 21000
766
+ },
767
+ {
768
+ "epoch": 2.59,
769
+ "grad_norm": 6.022315502166748,
770
+ "learning_rate": 4.741594516594517e-05,
771
+ "loss": 0.4005,
772
+ "step": 21500
773
+ },
774
+ {
775
+ "epoch": 2.59,
776
+ "eval_f1": 0.5169696711549415,
777
+ "eval_loss": 0.8356830477714539,
778
+ "eval_precision": 0.5518686564828851,
779
+ "eval_recall": 0.5099939487036261,
780
+ "eval_runtime": 9.9411,
781
+ "eval_samples_per_second": 88.32,
782
+ "eval_steps_per_second": 5.533,
783
+ "step": 21500
784
+ },
785
+ {
786
+ "epoch": 2.65,
787
+ "grad_norm": 3.7972123622894287,
788
+ "learning_rate": 4.735582010582011e-05,
789
+ "loss": 0.3907,
790
+ "step": 22000
791
+ },
792
+ {
793
+ "epoch": 2.65,
794
+ "eval_f1": 0.5552096784695296,
795
+ "eval_loss": 0.8016623258590698,
796
+ "eval_precision": 0.5884397258734974,
797
+ "eval_recall": 0.5408509053670344,
798
+ "eval_runtime": 9.5287,
799
+ "eval_samples_per_second": 92.142,
800
+ "eval_steps_per_second": 5.772,
801
+ "step": 22000
802
+ },
803
+ {
804
+ "epoch": 2.71,
805
+ "grad_norm": 6.792105674743652,
806
+ "learning_rate": 4.729569504569505e-05,
807
+ "loss": 0.3858,
808
+ "step": 22500
809
+ },
810
+ {
811
+ "epoch": 2.71,
812
+ "eval_f1": 0.5861875804431639,
813
+ "eval_loss": 0.8283036351203918,
814
+ "eval_precision": 0.6035754454158798,
815
+ "eval_recall": 0.5791807475678443,
816
+ "eval_runtime": 9.9189,
817
+ "eval_samples_per_second": 88.518,
818
+ "eval_steps_per_second": 5.545,
819
+ "step": 22500
820
+ },
821
+ {
822
+ "epoch": 2.77,
823
+ "grad_norm": 1.6373685598373413,
824
+ "learning_rate": 4.723556998556999e-05,
825
+ "loss": 0.3973,
826
+ "step": 23000
827
+ },
828
+ {
829
+ "epoch": 2.77,
830
+ "eval_f1": 0.5393179472998001,
831
+ "eval_loss": 0.9024349451065063,
832
+ "eval_precision": 0.5769620560542147,
833
+ "eval_recall": 0.5664646464646464,
834
+ "eval_runtime": 9.6142,
835
+ "eval_samples_per_second": 91.324,
836
+ "eval_steps_per_second": 5.721,
837
+ "step": 23000
838
+ },
839
+ {
840
+ "epoch": 2.83,
841
+ "grad_norm": 2.8107876777648926,
842
+ "learning_rate": 4.717544492544493e-05,
843
+ "loss": 0.3969,
844
+ "step": 23500
845
+ },
846
+ {
847
+ "epoch": 2.83,
848
+ "eval_f1": 0.5558204883761618,
849
+ "eval_loss": 0.8340879678726196,
850
+ "eval_precision": 0.564202659920544,
851
+ "eval_recall": 0.5527868547223386,
852
+ "eval_runtime": 9.7568,
853
+ "eval_samples_per_second": 89.988,
854
+ "eval_steps_per_second": 5.637,
855
+ "step": 23500
856
+ },
857
+ {
858
+ "epoch": 2.89,
859
+ "grad_norm": 3.1326379776000977,
860
+ "learning_rate": 4.7115319865319865e-05,
861
+ "loss": 0.3911,
862
+ "step": 24000
863
+ },
864
+ {
865
+ "epoch": 2.89,
866
+ "eval_f1": 0.5070306794950141,
867
+ "eval_loss": 0.8966326117515564,
868
+ "eval_precision": 0.6045140236228393,
869
+ "eval_recall": 0.5087874133035424,
870
+ "eval_runtime": 140.3321,
871
+ "eval_samples_per_second": 6.257,
872
+ "eval_steps_per_second": 0.392,
873
+ "step": 24000
874
+ },
875
+ {
876
+ "epoch": 2.95,
877
+ "grad_norm": 4.015880107879639,
878
+ "learning_rate": 4.7055194805194805e-05,
879
+ "loss": 0.3856,
880
+ "step": 24500
881
+ },
882
+ {
883
+ "epoch": 2.95,
884
+ "eval_f1": 0.5688512961191111,
885
+ "eval_loss": 0.8348873853683472,
886
+ "eval_precision": 0.6021456496959962,
887
+ "eval_recall": 0.5586072708653353,
888
+ "eval_runtime": 77.3964,
889
+ "eval_samples_per_second": 11.344,
890
+ "eval_steps_per_second": 0.711,
891
+ "step": 24500
892
+ },
893
+ {
894
+ "epoch": 3.01,
895
+ "grad_norm": 10.381326675415039,
896
+ "learning_rate": 4.699518999519e-05,
897
+ "loss": 0.3961,
898
+ "step": 25000
899
+ },
900
+ {
901
+ "epoch": 3.01,
902
+ "eval_f1": 0.558529842992594,
903
+ "eval_loss": 0.936424195766449,
904
+ "eval_precision": 0.6119355440337201,
905
+ "eval_recall": 0.5411525392170553,
906
+ "eval_runtime": 9.7898,
907
+ "eval_samples_per_second": 89.685,
908
+ "eval_steps_per_second": 5.618,
909
+ "step": 25000
910
+ },
911
+ {
912
+ "epoch": 3.07,
913
+ "grad_norm": 4.446761608123779,
914
+ "learning_rate": 4.693506493506494e-05,
915
+ "loss": 0.3301,
916
+ "step": 25500
917
+ },
918
+ {
919
+ "epoch": 3.07,
920
+ "eval_f1": 0.5812822447385155,
921
+ "eval_loss": 0.9541723728179932,
922
+ "eval_precision": 0.5757327569674934,
923
+ "eval_recall": 0.6084206116464181,
924
+ "eval_runtime": 9.902,
925
+ "eval_samples_per_second": 88.669,
926
+ "eval_steps_per_second": 5.554,
927
+ "step": 25500
928
+ },
929
+ {
930
+ "epoch": 3.13,
931
+ "grad_norm": 4.501039981842041,
932
+ "learning_rate": 4.687493987493988e-05,
933
+ "loss": 0.3385,
934
+ "step": 26000
935
+ },
936
+ {
937
+ "epoch": 3.13,
938
+ "eval_f1": 0.5345940260961672,
939
+ "eval_loss": 1.0137349367141724,
940
+ "eval_precision": 0.5563381916292068,
941
+ "eval_recall": 0.5293767164734907,
942
+ "eval_runtime": 10.1055,
943
+ "eval_samples_per_second": 86.883,
944
+ "eval_steps_per_second": 5.443,
945
+ "step": 26000
946
+ },
947
+ {
948
+ "epoch": 3.19,
949
+ "grad_norm": 1.9400558471679688,
950
+ "learning_rate": 4.681481481481482e-05,
951
+ "loss": 0.3475,
952
+ "step": 26500
953
+ },
954
+ {
955
+ "epoch": 3.19,
956
+ "eval_f1": 0.5822160079727499,
957
+ "eval_loss": 0.9310874938964844,
958
+ "eval_precision": 0.6358738260592424,
959
+ "eval_recall": 0.5675203649397198,
960
+ "eval_runtime": 9.9305,
961
+ "eval_samples_per_second": 88.414,
962
+ "eval_steps_per_second": 5.538,
963
+ "step": 26500
964
+ },
965
+ {
966
+ "epoch": 3.19,
967
+ "step": 26500,
968
+ "total_flos": 0.0,
969
+ "train_loss": 0.44209005161501325,
970
+ "train_runtime": 9420.7211,
971
+ "train_samples_per_second": 706.135,
972
+ "train_steps_per_second": 44.137
973
+ }
974
+ ],
975
+ "logging_steps": 500,
976
+ "max_steps": 415800,
977
+ "num_input_tokens_seen": 0,
978
+ "num_train_epochs": 50,
979
+ "save_steps": 500,
980
+ "total_flos": 0.0,
981
+ "train_batch_size": 16,
982
+ "trial_name": null,
983
+ "trial_params": null
984
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:395e85f2911b5fc0d75a2f29579390488ea7028271e65f7acc3187d8dd18ec45
3
+ size 4920