Samuel J. Huskey
commited on
Commit
·
9ccfaf8
1
Parent(s):
679d600
add: model files
Browse files- .gitignore +1 -0
- config.json +34 -0
- greek-latin-emissions.csv +2 -0
- logs/events.out.tfevents.1735139166.49c9ca38522d.6642.0 +3 -0
- logs/events.out.tfevents.1735139214.49c9ca38522d.7518.0 +3 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +7 -0
- tokenizer.json +0 -0
- tokenizer_config.json +56 -0
- trainer_state.json +1109 -0
- training_args.bin +3 -0
- vocab.txt +0 -0
.gitignore
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
*.DS_Store
|
config.json
ADDED
|
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "distilbert_multilingual_cased_greek_latin_classifier",
|
| 3 |
+
"activation": "gelu",
|
| 4 |
+
"architectures": [
|
| 5 |
+
"DistilBertForSequenceClassification"
|
| 6 |
+
],
|
| 7 |
+
"attention_dropout": 0.1,
|
| 8 |
+
"dim": 768,
|
| 9 |
+
"dropout": 0.1,
|
| 10 |
+
"hidden_dim": 3072,
|
| 11 |
+
"id2label": {
|
| 12 |
+
"0": "Greek",
|
| 13 |
+
"1": "Latin"
|
| 14 |
+
},
|
| 15 |
+
"initializer_range": 0.02,
|
| 16 |
+
"label2id": {
|
| 17 |
+
"Greek": 0,
|
| 18 |
+
"Latin": 1
|
| 19 |
+
},
|
| 20 |
+
"max_position_embeddings": 512,
|
| 21 |
+
"model_type": "distilbert",
|
| 22 |
+
"n_heads": 12,
|
| 23 |
+
"n_layers": 6,
|
| 24 |
+
"output_past": true,
|
| 25 |
+
"pad_token_id": 0,
|
| 26 |
+
"problem_type": "single_label_classification",
|
| 27 |
+
"qa_dropout": 0.1,
|
| 28 |
+
"seq_classif_dropout": 0.2,
|
| 29 |
+
"sinusoidal_pos_embds": false,
|
| 30 |
+
"tie_weights_": true,
|
| 31 |
+
"torch_dtype": "float32",
|
| 32 |
+
"transformers_version": "4.44.1",
|
| 33 |
+
"vocab_size": 119547
|
| 34 |
+
}
|
greek-latin-emissions.csv
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
timestamp,project_name,run_id,experiment_id,duration,emissions,emissions_rate,cpu_power,gpu_power,ram_power,cpu_energy,gpu_energy,ram_energy,energy_consumed,country_name,country_iso_code,region,cloud_provider,cloud_region,os,python_version,codecarbon_version,cpu_count,cpu_model,gpu_count,gpu_model,longitude,latitude,ram_total_size,tracking_mode,on_cloud,pue
|
| 2 |
+
2024-12-25T15:16:45,codecarbon,f658b237-20c1-45cf-a8ee-cdb5e8521351,5b0fa12a-3dd7-45bb-9766-cc326314d9f1,591.3640720729998,0.017871522337723503,3.0220845637570942e-05,42.5,105.04535891243202,31.30389261245728,0.006974275640934743,0.025850867347344004,0.005136130757212294,0.03796127374549103,Singapore,SGP,,,,Linux-6.1.85+-x86_64-with-glibc2.35,3.10.12,2.8.2,12,Intel(R) Xeon(R) CPU @ 2.20GHz,1,1 x NVIDIA A100-SXM4-40GB,103.8503,1.2868,83.47704696655273,machine,N,1.0
|
logs/events.out.tfevents.1735139166.49c9ca38522d.6642.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ec7414fa22502f5ef0d7c5f52d365638b066dd1919a6b3a93629d482d49a2c66
|
| 3 |
+
size 4184
|
logs/events.out.tfevents.1735139214.49c9ca38522d.7518.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0f838b6e8c8504aef5e7c251628fabfd9a900cf61dc48cc2d905e5a6564c1595
|
| 3 |
+
size 50378
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2c9296003fd0fb8618db1ed16d87b534346cc0ac4b6124cff8ed59299c5a04e3
|
| 3 |
+
size 541317368
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:21941a651fa3476118e1cdf74e425cc957ab413ac16623c4e0c6b8f4c7b5230f
|
| 3 |
+
size 1082696890
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f87605e6c67d5a5d9a11aa1efb02af19902881be7fdf3b3e1c6bbca0fd808e5c
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6f311cf2c6b954a71898a0497794c30a06bad24230f99e1c5541cecdc5621e10
|
| 3 |
+
size 1064
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": "[CLS]",
|
| 3 |
+
"mask_token": "[MASK]",
|
| 4 |
+
"pad_token": "[PAD]",
|
| 5 |
+
"sep_token": "[SEP]",
|
| 6 |
+
"unk_token": "[UNK]"
|
| 7 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[PAD]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"100": {
|
| 12 |
+
"content": "[UNK]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"101": {
|
| 20 |
+
"content": "[CLS]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"102": {
|
| 28 |
+
"content": "[SEP]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"103": {
|
| 36 |
+
"content": "[MASK]",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"clean_up_tokenization_spaces": false,
|
| 45 |
+
"cls_token": "[CLS]",
|
| 46 |
+
"do_lower_case": false,
|
| 47 |
+
"extra_special_tokens": {},
|
| 48 |
+
"mask_token": "[MASK]",
|
| 49 |
+
"model_max_length": 512,
|
| 50 |
+
"pad_token": "[PAD]",
|
| 51 |
+
"sep_token": "[SEP]",
|
| 52 |
+
"strip_accents": null,
|
| 53 |
+
"tokenize_chinese_chars": true,
|
| 54 |
+
"tokenizer_class": "DistilBertTokenizer",
|
| 55 |
+
"unk_token": "[UNK]"
|
| 56 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1109 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9635820582698403,
|
| 3 |
+
"best_model_checkpoint": "./distilbert_multilingual_cased_greek_latin_classifiergreek/checkpoint-14160",
|
| 4 |
+
"epoch": 8.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 14160,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.05649717514124294,
|
| 13 |
+
"grad_norm": 9.908761978149414,
|
| 14 |
+
"learning_rate": 4.9858757062146896e-05,
|
| 15 |
+
"loss": 0.4326,
|
| 16 |
+
"step": 100
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.11299435028248588,
|
| 20 |
+
"grad_norm": 0.47730961441993713,
|
| 21 |
+
"learning_rate": 4.971751412429379e-05,
|
| 22 |
+
"loss": 0.2915,
|
| 23 |
+
"step": 200
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.1694915254237288,
|
| 27 |
+
"grad_norm": 5.969715595245361,
|
| 28 |
+
"learning_rate": 4.957627118644068e-05,
|
| 29 |
+
"loss": 0.2426,
|
| 30 |
+
"step": 300
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.22598870056497175,
|
| 34 |
+
"grad_norm": 2.764862537384033,
|
| 35 |
+
"learning_rate": 4.9435028248587575e-05,
|
| 36 |
+
"loss": 0.256,
|
| 37 |
+
"step": 400
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.2824858757062147,
|
| 41 |
+
"grad_norm": 12.726994514465332,
|
| 42 |
+
"learning_rate": 4.929378531073446e-05,
|
| 43 |
+
"loss": 0.2165,
|
| 44 |
+
"step": 500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.3389830508474576,
|
| 48 |
+
"grad_norm": 1.1394743919372559,
|
| 49 |
+
"learning_rate": 4.915254237288136e-05,
|
| 50 |
+
"loss": 0.2168,
|
| 51 |
+
"step": 600
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.3954802259887006,
|
| 55 |
+
"grad_norm": 4.321178913116455,
|
| 56 |
+
"learning_rate": 4.9011299435028255e-05,
|
| 57 |
+
"loss": 0.2059,
|
| 58 |
+
"step": 700
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.4519774011299435,
|
| 62 |
+
"grad_norm": 18.238351821899414,
|
| 63 |
+
"learning_rate": 4.887005649717514e-05,
|
| 64 |
+
"loss": 0.2294,
|
| 65 |
+
"step": 800
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.5084745762711864,
|
| 69 |
+
"grad_norm": 7.210486888885498,
|
| 70 |
+
"learning_rate": 4.8728813559322034e-05,
|
| 71 |
+
"loss": 0.1904,
|
| 72 |
+
"step": 900
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.5649717514124294,
|
| 76 |
+
"grad_norm": 1.240628719329834,
|
| 77 |
+
"learning_rate": 4.8587570621468934e-05,
|
| 78 |
+
"loss": 0.1668,
|
| 79 |
+
"step": 1000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.6214689265536724,
|
| 83 |
+
"grad_norm": 9.029092788696289,
|
| 84 |
+
"learning_rate": 4.844632768361582e-05,
|
| 85 |
+
"loss": 0.18,
|
| 86 |
+
"step": 1100
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.6779661016949152,
|
| 90 |
+
"grad_norm": 4.866886615753174,
|
| 91 |
+
"learning_rate": 4.8305084745762714e-05,
|
| 92 |
+
"loss": 0.2082,
|
| 93 |
+
"step": 1200
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.7344632768361582,
|
| 97 |
+
"grad_norm": 7.6279778480529785,
|
| 98 |
+
"learning_rate": 4.816384180790961e-05,
|
| 99 |
+
"loss": 0.1818,
|
| 100 |
+
"step": 1300
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.7909604519774012,
|
| 104 |
+
"grad_norm": 8.820233345031738,
|
| 105 |
+
"learning_rate": 4.80225988700565e-05,
|
| 106 |
+
"loss": 0.1853,
|
| 107 |
+
"step": 1400
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.847457627118644,
|
| 111 |
+
"grad_norm": 22.571056365966797,
|
| 112 |
+
"learning_rate": 4.788135593220339e-05,
|
| 113 |
+
"loss": 0.1638,
|
| 114 |
+
"step": 1500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.903954802259887,
|
| 118 |
+
"grad_norm": 4.86266565322876,
|
| 119 |
+
"learning_rate": 4.7740112994350286e-05,
|
| 120 |
+
"loss": 0.1761,
|
| 121 |
+
"step": 1600
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.96045197740113,
|
| 125 |
+
"grad_norm": 2.2481741905212402,
|
| 126 |
+
"learning_rate": 4.759887005649718e-05,
|
| 127 |
+
"loss": 0.1481,
|
| 128 |
+
"step": 1700
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 1.0,
|
| 132 |
+
"eval_accuracy": 0.9473554127533367,
|
| 133 |
+
"eval_f1": 0.9471867274372374,
|
| 134 |
+
"eval_loss": 0.15506704151630402,
|
| 135 |
+
"eval_runtime": 0.99,
|
| 136 |
+
"eval_samples_per_second": 4086.911,
|
| 137 |
+
"eval_steps_per_second": 64.647,
|
| 138 |
+
"step": 1770
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 1.0169491525423728,
|
| 142 |
+
"grad_norm": 0.16318374872207642,
|
| 143 |
+
"learning_rate": 4.745762711864407e-05,
|
| 144 |
+
"loss": 0.1303,
|
| 145 |
+
"step": 1800
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 1.073446327683616,
|
| 149 |
+
"grad_norm": 2.8574020862579346,
|
| 150 |
+
"learning_rate": 4.7316384180790966e-05,
|
| 151 |
+
"loss": 0.0916,
|
| 152 |
+
"step": 1900
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 1.1299435028248588,
|
| 156 |
+
"grad_norm": 7.220451831817627,
|
| 157 |
+
"learning_rate": 4.717514124293785e-05,
|
| 158 |
+
"loss": 0.1214,
|
| 159 |
+
"step": 2000
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 1.1864406779661016,
|
| 163 |
+
"grad_norm": 0.23234207928180695,
|
| 164 |
+
"learning_rate": 4.703389830508475e-05,
|
| 165 |
+
"loss": 0.1005,
|
| 166 |
+
"step": 2100
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 1.2429378531073447,
|
| 170 |
+
"grad_norm": 5.788926124572754,
|
| 171 |
+
"learning_rate": 4.689265536723164e-05,
|
| 172 |
+
"loss": 0.106,
|
| 173 |
+
"step": 2200
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 1.2994350282485876,
|
| 177 |
+
"grad_norm": 14.127985954284668,
|
| 178 |
+
"learning_rate": 4.675141242937853e-05,
|
| 179 |
+
"loss": 0.1002,
|
| 180 |
+
"step": 2300
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 1.3559322033898304,
|
| 184 |
+
"grad_norm": 8.628386497497559,
|
| 185 |
+
"learning_rate": 4.6610169491525425e-05,
|
| 186 |
+
"loss": 0.1228,
|
| 187 |
+
"step": 2400
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 1.4124293785310735,
|
| 191 |
+
"grad_norm": 3.4723896980285645,
|
| 192 |
+
"learning_rate": 4.646892655367232e-05,
|
| 193 |
+
"loss": 0.1193,
|
| 194 |
+
"step": 2500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 1.4689265536723164,
|
| 198 |
+
"grad_norm": 5.112296104431152,
|
| 199 |
+
"learning_rate": 4.632768361581921e-05,
|
| 200 |
+
"loss": 0.0989,
|
| 201 |
+
"step": 2600
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 1.5254237288135593,
|
| 205 |
+
"grad_norm": 8.718145370483398,
|
| 206 |
+
"learning_rate": 4.6186440677966104e-05,
|
| 207 |
+
"loss": 0.1291,
|
| 208 |
+
"step": 2700
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 1.5819209039548023,
|
| 212 |
+
"grad_norm": 0.052204638719558716,
|
| 213 |
+
"learning_rate": 4.6045197740113e-05,
|
| 214 |
+
"loss": 0.0987,
|
| 215 |
+
"step": 2800
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 1.6384180790960452,
|
| 219 |
+
"grad_norm": 11.152572631835938,
|
| 220 |
+
"learning_rate": 4.590395480225989e-05,
|
| 221 |
+
"loss": 0.113,
|
| 222 |
+
"step": 2900
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 1.694915254237288,
|
| 226 |
+
"grad_norm": 3.002537965774536,
|
| 227 |
+
"learning_rate": 4.5762711864406784e-05,
|
| 228 |
+
"loss": 0.1215,
|
| 229 |
+
"step": 3000
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 1.7514124293785311,
|
| 233 |
+
"grad_norm": 16.578323364257812,
|
| 234 |
+
"learning_rate": 4.562146892655367e-05,
|
| 235 |
+
"loss": 0.0998,
|
| 236 |
+
"step": 3100
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 1.807909604519774,
|
| 240 |
+
"grad_norm": 4.660722255706787,
|
| 241 |
+
"learning_rate": 4.548022598870056e-05,
|
| 242 |
+
"loss": 0.1015,
|
| 243 |
+
"step": 3200
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 1.8644067796610169,
|
| 247 |
+
"grad_norm": 0.32472193241119385,
|
| 248 |
+
"learning_rate": 4.533898305084746e-05,
|
| 249 |
+
"loss": 0.0982,
|
| 250 |
+
"step": 3300
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 1.92090395480226,
|
| 254 |
+
"grad_norm": 12.544636726379395,
|
| 255 |
+
"learning_rate": 4.519774011299435e-05,
|
| 256 |
+
"loss": 0.1151,
|
| 257 |
+
"step": 3400
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 1.9774011299435028,
|
| 261 |
+
"grad_norm": 0.20591090619564056,
|
| 262 |
+
"learning_rate": 4.505649717514124e-05,
|
| 263 |
+
"loss": 0.1109,
|
| 264 |
+
"step": 3500
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 2.0,
|
| 268 |
+
"eval_accuracy": 0.9540286702916461,
|
| 269 |
+
"eval_f1": 0.9542995811514169,
|
| 270 |
+
"eval_loss": 0.17088210582733154,
|
| 271 |
+
"eval_runtime": 0.9919,
|
| 272 |
+
"eval_samples_per_second": 4078.94,
|
| 273 |
+
"eval_steps_per_second": 64.521,
|
| 274 |
+
"step": 3540
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 2.0338983050847457,
|
| 278 |
+
"grad_norm": 1.156205177307129,
|
| 279 |
+
"learning_rate": 4.491525423728814e-05,
|
| 280 |
+
"loss": 0.0764,
|
| 281 |
+
"step": 3600
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 2.0903954802259888,
|
| 285 |
+
"grad_norm": 4.644138336181641,
|
| 286 |
+
"learning_rate": 4.477401129943503e-05,
|
| 287 |
+
"loss": 0.0603,
|
| 288 |
+
"step": 3700
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 2.146892655367232,
|
| 292 |
+
"grad_norm": 0.0709792822599411,
|
| 293 |
+
"learning_rate": 4.463276836158192e-05,
|
| 294 |
+
"loss": 0.0649,
|
| 295 |
+
"step": 3800
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 2.2033898305084745,
|
| 299 |
+
"grad_norm": 0.13090333342552185,
|
| 300 |
+
"learning_rate": 4.4491525423728816e-05,
|
| 301 |
+
"loss": 0.076,
|
| 302 |
+
"step": 3900
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 2.2598870056497176,
|
| 306 |
+
"grad_norm": 39.72850799560547,
|
| 307 |
+
"learning_rate": 4.435028248587571e-05,
|
| 308 |
+
"loss": 0.0827,
|
| 309 |
+
"step": 4000
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 2.3163841807909606,
|
| 313 |
+
"grad_norm": 0.29564905166625977,
|
| 314 |
+
"learning_rate": 4.42090395480226e-05,
|
| 315 |
+
"loss": 0.0701,
|
| 316 |
+
"step": 4100
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 2.3728813559322033,
|
| 320 |
+
"grad_norm": 0.23284725844860077,
|
| 321 |
+
"learning_rate": 4.4067796610169495e-05,
|
| 322 |
+
"loss": 0.0697,
|
| 323 |
+
"step": 4200
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 2.4293785310734464,
|
| 327 |
+
"grad_norm": 0.059655264019966125,
|
| 328 |
+
"learning_rate": 4.392655367231638e-05,
|
| 329 |
+
"loss": 0.0803,
|
| 330 |
+
"step": 4300
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 2.4858757062146895,
|
| 334 |
+
"grad_norm": 0.01876319944858551,
|
| 335 |
+
"learning_rate": 4.378531073446328e-05,
|
| 336 |
+
"loss": 0.0796,
|
| 337 |
+
"step": 4400
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 2.542372881355932,
|
| 341 |
+
"grad_norm": 0.07363492995500565,
|
| 342 |
+
"learning_rate": 4.3644067796610175e-05,
|
| 343 |
+
"loss": 0.0573,
|
| 344 |
+
"step": 4500
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 2.598870056497175,
|
| 348 |
+
"grad_norm": 0.26911139488220215,
|
| 349 |
+
"learning_rate": 4.350282485875706e-05,
|
| 350 |
+
"loss": 0.0732,
|
| 351 |
+
"step": 4600
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 2.655367231638418,
|
| 355 |
+
"grad_norm": 0.045297879725694656,
|
| 356 |
+
"learning_rate": 4.3361581920903954e-05,
|
| 357 |
+
"loss": 0.0645,
|
| 358 |
+
"step": 4700
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 2.711864406779661,
|
| 362 |
+
"grad_norm": 0.24285119771957397,
|
| 363 |
+
"learning_rate": 4.3220338983050854e-05,
|
| 364 |
+
"loss": 0.0852,
|
| 365 |
+
"step": 4800
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 2.768361581920904,
|
| 369 |
+
"grad_norm": 70.39765930175781,
|
| 370 |
+
"learning_rate": 4.307909604519774e-05,
|
| 371 |
+
"loss": 0.0763,
|
| 372 |
+
"step": 4900
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 2.824858757062147,
|
| 376 |
+
"grad_norm": 3.1160919666290283,
|
| 377 |
+
"learning_rate": 4.2937853107344634e-05,
|
| 378 |
+
"loss": 0.0618,
|
| 379 |
+
"step": 5000
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 2.8813559322033897,
|
| 383 |
+
"grad_norm": 0.14466217160224915,
|
| 384 |
+
"learning_rate": 4.279661016949153e-05,
|
| 385 |
+
"loss": 0.0628,
|
| 386 |
+
"step": 5100
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 2.937853107344633,
|
| 390 |
+
"grad_norm": 11.605415344238281,
|
| 391 |
+
"learning_rate": 4.265536723163842e-05,
|
| 392 |
+
"loss": 0.0539,
|
| 393 |
+
"step": 5200
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 2.994350282485876,
|
| 397 |
+
"grad_norm": 0.03261380270123482,
|
| 398 |
+
"learning_rate": 4.251412429378531e-05,
|
| 399 |
+
"loss": 0.0349,
|
| 400 |
+
"step": 5300
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 3.0,
|
| 404 |
+
"eval_accuracy": 0.9574888779041029,
|
| 405 |
+
"eval_f1": 0.9573208648613947,
|
| 406 |
+
"eval_loss": 0.20843710005283356,
|
| 407 |
+
"eval_runtime": 0.9688,
|
| 408 |
+
"eval_samples_per_second": 4176.286,
|
| 409 |
+
"eval_steps_per_second": 66.061,
|
| 410 |
+
"step": 5310
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 3.0508474576271185,
|
| 414 |
+
"grad_norm": 0.007287267595529556,
|
| 415 |
+
"learning_rate": 4.2372881355932206e-05,
|
| 416 |
+
"loss": 0.0575,
|
| 417 |
+
"step": 5400
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 3.1073446327683616,
|
| 421 |
+
"grad_norm": 19.718822479248047,
|
| 422 |
+
"learning_rate": 4.22316384180791e-05,
|
| 423 |
+
"loss": 0.0423,
|
| 424 |
+
"step": 5500
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 3.1638418079096047,
|
| 428 |
+
"grad_norm": 78.35333251953125,
|
| 429 |
+
"learning_rate": 4.209039548022599e-05,
|
| 430 |
+
"loss": 0.0349,
|
| 431 |
+
"step": 5600
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 3.2203389830508473,
|
| 435 |
+
"grad_norm": 0.025877630338072777,
|
| 436 |
+
"learning_rate": 4.1949152542372886e-05,
|
| 437 |
+
"loss": 0.0632,
|
| 438 |
+
"step": 5700
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 3.2768361581920904,
|
| 442 |
+
"grad_norm": 14.864492416381836,
|
| 443 |
+
"learning_rate": 4.180790960451977e-05,
|
| 444 |
+
"loss": 0.0651,
|
| 445 |
+
"step": 5800
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 3.3333333333333335,
|
| 449 |
+
"grad_norm": 0.002800008049234748,
|
| 450 |
+
"learning_rate": 4.166666666666667e-05,
|
| 451 |
+
"loss": 0.0225,
|
| 452 |
+
"step": 5900
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 3.389830508474576,
|
| 456 |
+
"grad_norm": 0.10183978080749512,
|
| 457 |
+
"learning_rate": 4.152542372881356e-05,
|
| 458 |
+
"loss": 0.0467,
|
| 459 |
+
"step": 6000
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 3.446327683615819,
|
| 463 |
+
"grad_norm": 30.69606590270996,
|
| 464 |
+
"learning_rate": 4.138418079096045e-05,
|
| 465 |
+
"loss": 0.0509,
|
| 466 |
+
"step": 6100
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 3.5028248587570623,
|
| 470 |
+
"grad_norm": 0.43340635299682617,
|
| 471 |
+
"learning_rate": 4.1242937853107345e-05,
|
| 472 |
+
"loss": 0.0366,
|
| 473 |
+
"step": 6200
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 3.559322033898305,
|
| 477 |
+
"grad_norm": 3.5696895122528076,
|
| 478 |
+
"learning_rate": 4.110169491525424e-05,
|
| 479 |
+
"loss": 0.0576,
|
| 480 |
+
"step": 6300
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 3.615819209039548,
|
| 484 |
+
"grad_norm": 3.981534481048584,
|
| 485 |
+
"learning_rate": 4.096045197740113e-05,
|
| 486 |
+
"loss": 0.0574,
|
| 487 |
+
"step": 6400
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 3.672316384180791,
|
| 491 |
+
"grad_norm": 0.020425381138920784,
|
| 492 |
+
"learning_rate": 4.0819209039548024e-05,
|
| 493 |
+
"loss": 0.0582,
|
| 494 |
+
"step": 6500
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 3.7288135593220337,
|
| 498 |
+
"grad_norm": 1.6137280464172363,
|
| 499 |
+
"learning_rate": 4.067796610169492e-05,
|
| 500 |
+
"loss": 0.037,
|
| 501 |
+
"step": 6600
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 3.785310734463277,
|
| 505 |
+
"grad_norm": 0.41225990653038025,
|
| 506 |
+
"learning_rate": 4.053672316384181e-05,
|
| 507 |
+
"loss": 0.039,
|
| 508 |
+
"step": 6700
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 3.84180790960452,
|
| 512 |
+
"grad_norm": 0.02651926688849926,
|
| 513 |
+
"learning_rate": 4.0395480225988704e-05,
|
| 514 |
+
"loss": 0.052,
|
| 515 |
+
"step": 6800
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 3.898305084745763,
|
| 519 |
+
"grad_norm": 4.174577713012695,
|
| 520 |
+
"learning_rate": 4.025423728813559e-05,
|
| 521 |
+
"loss": 0.0746,
|
| 522 |
+
"step": 6900
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 3.9548022598870056,
|
| 526 |
+
"grad_norm": 0.06549729406833649,
|
| 527 |
+
"learning_rate": 4.011299435028249e-05,
|
| 528 |
+
"loss": 0.0682,
|
| 529 |
+
"step": 7000
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 4.0,
|
| 533 |
+
"eval_accuracy": 0.9619377162629758,
|
| 534 |
+
"eval_f1": 0.961922353652765,
|
| 535 |
+
"eval_loss": 0.19074885547161102,
|
| 536 |
+
"eval_runtime": 0.9577,
|
| 537 |
+
"eval_samples_per_second": 4224.832,
|
| 538 |
+
"eval_steps_per_second": 66.829,
|
| 539 |
+
"step": 7080
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 4.011299435028248,
|
| 543 |
+
"grad_norm": 0.09852942079305649,
|
| 544 |
+
"learning_rate": 3.997175141242938e-05,
|
| 545 |
+
"loss": 0.0573,
|
| 546 |
+
"step": 7100
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 4.067796610169491,
|
| 550 |
+
"grad_norm": 0.010253222659230232,
|
| 551 |
+
"learning_rate": 3.983050847457627e-05,
|
| 552 |
+
"loss": 0.0268,
|
| 553 |
+
"step": 7200
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 4.124293785310734,
|
| 557 |
+
"grad_norm": 0.05561167746782303,
|
| 558 |
+
"learning_rate": 3.968926553672316e-05,
|
| 559 |
+
"loss": 0.041,
|
| 560 |
+
"step": 7300
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 4.1807909604519775,
|
| 564 |
+
"grad_norm": 0.020777329802513123,
|
| 565 |
+
"learning_rate": 3.954802259887006e-05,
|
| 566 |
+
"loss": 0.0428,
|
| 567 |
+
"step": 7400
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 4.237288135593221,
|
| 571 |
+
"grad_norm": 0.011439072899520397,
|
| 572 |
+
"learning_rate": 3.940677966101695e-05,
|
| 573 |
+
"loss": 0.0281,
|
| 574 |
+
"step": 7500
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 4.293785310734464,
|
| 578 |
+
"grad_norm": 0.29063406586647034,
|
| 579 |
+
"learning_rate": 3.926553672316384e-05,
|
| 580 |
+
"loss": 0.0347,
|
| 581 |
+
"step": 7600
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 4.350282485875706,
|
| 585 |
+
"grad_norm": 0.008078676648437977,
|
| 586 |
+
"learning_rate": 3.9124293785310735e-05,
|
| 587 |
+
"loss": 0.0427,
|
| 588 |
+
"step": 7700
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 4.406779661016949,
|
| 592 |
+
"grad_norm": 0.0378178134560585,
|
| 593 |
+
"learning_rate": 3.898305084745763e-05,
|
| 594 |
+
"loss": 0.0448,
|
| 595 |
+
"step": 7800
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 4.463276836158192,
|
| 599 |
+
"grad_norm": 0.09035930037498474,
|
| 600 |
+
"learning_rate": 3.884180790960452e-05,
|
| 601 |
+
"loss": 0.0328,
|
| 602 |
+
"step": 7900
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 4.519774011299435,
|
| 606 |
+
"grad_norm": 0.05402543023228645,
|
| 607 |
+
"learning_rate": 3.8700564971751415e-05,
|
| 608 |
+
"loss": 0.0345,
|
| 609 |
+
"step": 8000
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 4.576271186440678,
|
| 613 |
+
"grad_norm": 0.01713019795715809,
|
| 614 |
+
"learning_rate": 3.855932203389831e-05,
|
| 615 |
+
"loss": 0.0358,
|
| 616 |
+
"step": 8100
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 4.632768361581921,
|
| 620 |
+
"grad_norm": 0.0475781112909317,
|
| 621 |
+
"learning_rate": 3.84180790960452e-05,
|
| 622 |
+
"loss": 0.0532,
|
| 623 |
+
"step": 8200
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 4.6892655367231635,
|
| 627 |
+
"grad_norm": 0.006405588239431381,
|
| 628 |
+
"learning_rate": 3.8276836158192094e-05,
|
| 629 |
+
"loss": 0.0324,
|
| 630 |
+
"step": 8300
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 4.745762711864407,
|
| 634 |
+
"grad_norm": 4.002650260925293,
|
| 635 |
+
"learning_rate": 3.813559322033898e-05,
|
| 636 |
+
"loss": 0.0526,
|
| 637 |
+
"step": 8400
|
| 638 |
+
},
|
| 639 |
+
{
|
| 640 |
+
"epoch": 4.80225988700565,
|
| 641 |
+
"grad_norm": 0.010295218788087368,
|
| 642 |
+
"learning_rate": 3.799435028248588e-05,
|
| 643 |
+
"loss": 0.0428,
|
| 644 |
+
"step": 8500
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 4.858757062146893,
|
| 648 |
+
"grad_norm": 25.15464973449707,
|
| 649 |
+
"learning_rate": 3.7853107344632774e-05,
|
| 650 |
+
"loss": 0.0513,
|
| 651 |
+
"step": 8600
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 4.915254237288136,
|
| 655 |
+
"grad_norm": 0.018476568162441254,
|
| 656 |
+
"learning_rate": 3.771186440677966e-05,
|
| 657 |
+
"loss": 0.0627,
|
| 658 |
+
"step": 8700
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 4.971751412429379,
|
| 662 |
+
"grad_norm": 0.02234013006091118,
|
| 663 |
+
"learning_rate": 3.7570621468926554e-05,
|
| 664 |
+
"loss": 0.0436,
|
| 665 |
+
"step": 8800
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 5.0,
|
| 669 |
+
"eval_accuracy": 0.9594661393969353,
|
| 670 |
+
"eval_f1": 0.9595618588245926,
|
| 671 |
+
"eval_loss": 0.24679133296012878,
|
| 672 |
+
"eval_runtime": 0.9625,
|
| 673 |
+
"eval_samples_per_second": 4203.719,
|
| 674 |
+
"eval_steps_per_second": 66.495,
|
| 675 |
+
"step": 8850
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 5.028248587570621,
|
| 679 |
+
"grad_norm": 0.007325501646846533,
|
| 680 |
+
"learning_rate": 3.7429378531073453e-05,
|
| 681 |
+
"loss": 0.011,
|
| 682 |
+
"step": 8900
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 5.084745762711864,
|
| 686 |
+
"grad_norm": 0.015004786662757397,
|
| 687 |
+
"learning_rate": 3.728813559322034e-05,
|
| 688 |
+
"loss": 0.0239,
|
| 689 |
+
"step": 9000
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 5.141242937853107,
|
| 693 |
+
"grad_norm": 0.02809782139956951,
|
| 694 |
+
"learning_rate": 3.714689265536723e-05,
|
| 695 |
+
"loss": 0.0468,
|
| 696 |
+
"step": 9100
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 5.19774011299435,
|
| 700 |
+
"grad_norm": 0.061971381306648254,
|
| 701 |
+
"learning_rate": 3.7005649717514126e-05,
|
| 702 |
+
"loss": 0.0369,
|
| 703 |
+
"step": 9200
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 5.254237288135593,
|
| 707 |
+
"grad_norm": 0.028554769232869148,
|
| 708 |
+
"learning_rate": 3.686440677966102e-05,
|
| 709 |
+
"loss": 0.0254,
|
| 710 |
+
"step": 9300
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 5.3107344632768365,
|
| 714 |
+
"grad_norm": 0.049820106476545334,
|
| 715 |
+
"learning_rate": 3.672316384180791e-05,
|
| 716 |
+
"loss": 0.0371,
|
| 717 |
+
"step": 9400
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 5.367231638418079,
|
| 721 |
+
"grad_norm": 0.016609592363238335,
|
| 722 |
+
"learning_rate": 3.6581920903954806e-05,
|
| 723 |
+
"loss": 0.0184,
|
| 724 |
+
"step": 9500
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 5.423728813559322,
|
| 728 |
+
"grad_norm": 0.05181876942515373,
|
| 729 |
+
"learning_rate": 3.644067796610169e-05,
|
| 730 |
+
"loss": 0.0414,
|
| 731 |
+
"step": 9600
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 5.480225988700565,
|
| 735 |
+
"grad_norm": 0.05821879953145981,
|
| 736 |
+
"learning_rate": 3.629943502824859e-05,
|
| 737 |
+
"loss": 0.0308,
|
| 738 |
+
"step": 9700
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 5.536723163841808,
|
| 742 |
+
"grad_norm": 0.010366985574364662,
|
| 743 |
+
"learning_rate": 3.6158192090395485e-05,
|
| 744 |
+
"loss": 0.0278,
|
| 745 |
+
"step": 9800
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 5.593220338983051,
|
| 749 |
+
"grad_norm": 0.019191740080714226,
|
| 750 |
+
"learning_rate": 3.601694915254237e-05,
|
| 751 |
+
"loss": 0.0435,
|
| 752 |
+
"step": 9900
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 5.649717514124294,
|
| 756 |
+
"grad_norm": 0.06532129645347595,
|
| 757 |
+
"learning_rate": 3.587570621468927e-05,
|
| 758 |
+
"loss": 0.0237,
|
| 759 |
+
"step": 10000
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 5.706214689265536,
|
| 763 |
+
"grad_norm": 0.009392981417477131,
|
| 764 |
+
"learning_rate": 3.573446327683616e-05,
|
| 765 |
+
"loss": 0.0334,
|
| 766 |
+
"step": 10100
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 5.762711864406779,
|
| 770 |
+
"grad_norm": 0.023171979933977127,
|
| 771 |
+
"learning_rate": 3.559322033898305e-05,
|
| 772 |
+
"loss": 0.0487,
|
| 773 |
+
"step": 10200
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 5.8192090395480225,
|
| 777 |
+
"grad_norm": 0.055124878883361816,
|
| 778 |
+
"learning_rate": 3.5451977401129944e-05,
|
| 779 |
+
"loss": 0.0412,
|
| 780 |
+
"step": 10300
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 5.875706214689266,
|
| 784 |
+
"grad_norm": 0.015424055978655815,
|
| 785 |
+
"learning_rate": 3.531073446327684e-05,
|
| 786 |
+
"loss": 0.0292,
|
| 787 |
+
"step": 10400
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 5.932203389830509,
|
| 791 |
+
"grad_norm": 0.6497403979301453,
|
| 792 |
+
"learning_rate": 3.516949152542373e-05,
|
| 793 |
+
"loss": 0.039,
|
| 794 |
+
"step": 10500
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 5.988700564971752,
|
| 798 |
+
"grad_norm": 18.98410415649414,
|
| 799 |
+
"learning_rate": 3.5028248587570624e-05,
|
| 800 |
+
"loss": 0.0322,
|
| 801 |
+
"step": 10600
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 6.0,
|
| 805 |
+
"eval_accuracy": 0.9614434008897677,
|
| 806 |
+
"eval_f1": 0.9615565973130906,
|
| 807 |
+
"eval_loss": 0.2411661297082901,
|
| 808 |
+
"eval_runtime": 0.9629,
|
| 809 |
+
"eval_samples_per_second": 4201.762,
|
| 810 |
+
"eval_steps_per_second": 66.464,
|
| 811 |
+
"step": 10620
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 6.045197740112994,
|
| 815 |
+
"grad_norm": 0.013564531691372395,
|
| 816 |
+
"learning_rate": 3.488700564971752e-05,
|
| 817 |
+
"loss": 0.0254,
|
| 818 |
+
"step": 10700
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 6.101694915254237,
|
| 822 |
+
"grad_norm": 33.3035888671875,
|
| 823 |
+
"learning_rate": 3.474576271186441e-05,
|
| 824 |
+
"loss": 0.0295,
|
| 825 |
+
"step": 10800
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 6.15819209039548,
|
| 829 |
+
"grad_norm": 0.12126260250806808,
|
| 830 |
+
"learning_rate": 3.46045197740113e-05,
|
| 831 |
+
"loss": 0.0275,
|
| 832 |
+
"step": 10900
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 6.214689265536723,
|
| 836 |
+
"grad_norm": 0.03739802539348602,
|
| 837 |
+
"learning_rate": 3.446327683615819e-05,
|
| 838 |
+
"loss": 0.0267,
|
| 839 |
+
"step": 11000
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 6.271186440677966,
|
| 843 |
+
"grad_norm": 0.03359340503811836,
|
| 844 |
+
"learning_rate": 3.432203389830508e-05,
|
| 845 |
+
"loss": 0.0389,
|
| 846 |
+
"step": 11100
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"epoch": 6.327683615819209,
|
| 850 |
+
"grad_norm": 0.003635927801951766,
|
| 851 |
+
"learning_rate": 3.418079096045198e-05,
|
| 852 |
+
"loss": 0.0255,
|
| 853 |
+
"step": 11200
|
| 854 |
+
},
|
| 855 |
+
{
|
| 856 |
+
"epoch": 6.3841807909604515,
|
| 857 |
+
"grad_norm": 0.06124364957213402,
|
| 858 |
+
"learning_rate": 3.403954802259887e-05,
|
| 859 |
+
"loss": 0.0229,
|
| 860 |
+
"step": 11300
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 6.440677966101695,
|
| 864 |
+
"grad_norm": 0.026170525699853897,
|
| 865 |
+
"learning_rate": 3.389830508474576e-05,
|
| 866 |
+
"loss": 0.0319,
|
| 867 |
+
"step": 11400
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 6.497175141242938,
|
| 871 |
+
"grad_norm": 0.013875061646103859,
|
| 872 |
+
"learning_rate": 3.375706214689266e-05,
|
| 873 |
+
"loss": 0.0276,
|
| 874 |
+
"step": 11500
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 6.553672316384181,
|
| 878 |
+
"grad_norm": 0.009600764140486717,
|
| 879 |
+
"learning_rate": 3.361581920903955e-05,
|
| 880 |
+
"loss": 0.0182,
|
| 881 |
+
"step": 11600
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 6.610169491525424,
|
| 885 |
+
"grad_norm": 0.02147483266890049,
|
| 886 |
+
"learning_rate": 3.347457627118644e-05,
|
| 887 |
+
"loss": 0.0277,
|
| 888 |
+
"step": 11700
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 6.666666666666667,
|
| 892 |
+
"grad_norm": 0.007301884237676859,
|
| 893 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 894 |
+
"loss": 0.0268,
|
| 895 |
+
"step": 11800
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 6.72316384180791,
|
| 899 |
+
"grad_norm": 0.008684027940034866,
|
| 900 |
+
"learning_rate": 3.319209039548023e-05,
|
| 901 |
+
"loss": 0.0244,
|
| 902 |
+
"step": 11900
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 6.779661016949152,
|
| 906 |
+
"grad_norm": 0.0058201453648507595,
|
| 907 |
+
"learning_rate": 3.305084745762712e-05,
|
| 908 |
+
"loss": 0.018,
|
| 909 |
+
"step": 12000
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 6.836158192090395,
|
| 913 |
+
"grad_norm": 0.015645477920770645,
|
| 914 |
+
"learning_rate": 3.2909604519774014e-05,
|
| 915 |
+
"loss": 0.0389,
|
| 916 |
+
"step": 12100
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 6.892655367231638,
|
| 920 |
+
"grad_norm": 0.013589623384177685,
|
| 921 |
+
"learning_rate": 3.27683615819209e-05,
|
| 922 |
+
"loss": 0.0271,
|
| 923 |
+
"step": 12200
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 6.9491525423728815,
|
| 927 |
+
"grad_norm": 0.004052096512168646,
|
| 928 |
+
"learning_rate": 3.26271186440678e-05,
|
| 929 |
+
"loss": 0.012,
|
| 930 |
+
"step": 12300
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 7.0,
|
| 934 |
+
"eval_accuracy": 0.963173504695996,
|
| 935 |
+
"eval_f1": 0.9631919351432553,
|
| 936 |
+
"eval_loss": 0.22085699439048767,
|
| 937 |
+
"eval_runtime": 0.9623,
|
| 938 |
+
"eval_samples_per_second": 4204.299,
|
| 939 |
+
"eval_steps_per_second": 66.504,
|
| 940 |
+
"step": 12390
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 7.005649717514125,
|
| 944 |
+
"grad_norm": 0.06705684214830399,
|
| 945 |
+
"learning_rate": 3.2485875706214694e-05,
|
| 946 |
+
"loss": 0.0309,
|
| 947 |
+
"step": 12400
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 7.062146892655368,
|
| 951 |
+
"grad_norm": 0.006240461952984333,
|
| 952 |
+
"learning_rate": 3.234463276836158e-05,
|
| 953 |
+
"loss": 0.0084,
|
| 954 |
+
"step": 12500
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 7.11864406779661,
|
| 958 |
+
"grad_norm": 0.020344626158475876,
|
| 959 |
+
"learning_rate": 3.2203389830508473e-05,
|
| 960 |
+
"loss": 0.0168,
|
| 961 |
+
"step": 12600
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 7.175141242937853,
|
| 965 |
+
"grad_norm": 0.003926662262529135,
|
| 966 |
+
"learning_rate": 3.2062146892655373e-05,
|
| 967 |
+
"loss": 0.022,
|
| 968 |
+
"step": 12700
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 7.231638418079096,
|
| 972 |
+
"grad_norm": 0.0025492089334875345,
|
| 973 |
+
"learning_rate": 3.192090395480226e-05,
|
| 974 |
+
"loss": 0.0133,
|
| 975 |
+
"step": 12800
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 7.288135593220339,
|
| 979 |
+
"grad_norm": 0.005623087752610445,
|
| 980 |
+
"learning_rate": 3.177966101694915e-05,
|
| 981 |
+
"loss": 0.0164,
|
| 982 |
+
"step": 12900
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 7.344632768361582,
|
| 986 |
+
"grad_norm": 0.0032459620852023363,
|
| 987 |
+
"learning_rate": 3.1638418079096046e-05,
|
| 988 |
+
"loss": 0.0272,
|
| 989 |
+
"step": 13000
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 7.401129943502825,
|
| 993 |
+
"grad_norm": 1.1293178796768188,
|
| 994 |
+
"learning_rate": 3.149717514124294e-05,
|
| 995 |
+
"loss": 0.0148,
|
| 996 |
+
"step": 13100
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 7.4576271186440675,
|
| 1000 |
+
"grad_norm": 0.0017996145179495215,
|
| 1001 |
+
"learning_rate": 3.135593220338983e-05,
|
| 1002 |
+
"loss": 0.0132,
|
| 1003 |
+
"step": 13200
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 7.5141242937853105,
|
| 1007 |
+
"grad_norm": 0.008758709765970707,
|
| 1008 |
+
"learning_rate": 3.1214689265536726e-05,
|
| 1009 |
+
"loss": 0.0152,
|
| 1010 |
+
"step": 13300
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 7.570621468926554,
|
| 1014 |
+
"grad_norm": 0.0038798090536147356,
|
| 1015 |
+
"learning_rate": 3.107344632768362e-05,
|
| 1016 |
+
"loss": 0.0106,
|
| 1017 |
+
"step": 13400
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 7.627118644067797,
|
| 1021 |
+
"grad_norm": 0.005076746456325054,
|
| 1022 |
+
"learning_rate": 3.093220338983051e-05,
|
| 1023 |
+
"loss": 0.0158,
|
| 1024 |
+
"step": 13500
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 7.68361581920904,
|
| 1028 |
+
"grad_norm": 0.003670661011710763,
|
| 1029 |
+
"learning_rate": 3.0790960451977405e-05,
|
| 1030 |
+
"loss": 0.0093,
|
| 1031 |
+
"step": 13600
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 7.740112994350282,
|
| 1035 |
+
"grad_norm": 0.003522429848089814,
|
| 1036 |
+
"learning_rate": 3.064971751412429e-05,
|
| 1037 |
+
"loss": 0.0183,
|
| 1038 |
+
"step": 13700
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 7.796610169491525,
|
| 1042 |
+
"grad_norm": 0.06700780242681503,
|
| 1043 |
+
"learning_rate": 3.050847457627119e-05,
|
| 1044 |
+
"loss": 0.0398,
|
| 1045 |
+
"step": 13800
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 7.853107344632768,
|
| 1049 |
+
"grad_norm": 0.01462018396705389,
|
| 1050 |
+
"learning_rate": 3.036723163841808e-05,
|
| 1051 |
+
"loss": 0.0286,
|
| 1052 |
+
"step": 13900
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 7.909604519774011,
|
| 1056 |
+
"grad_norm": 0.025290269404649734,
|
| 1057 |
+
"learning_rate": 3.022598870056497e-05,
|
| 1058 |
+
"loss": 0.037,
|
| 1059 |
+
"step": 14000
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 7.966101694915254,
|
| 1063 |
+
"grad_norm": 0.011192042380571365,
|
| 1064 |
+
"learning_rate": 3.0084745762711864e-05,
|
| 1065 |
+
"loss": 0.0147,
|
| 1066 |
+
"step": 14100
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 8.0,
|
| 1070 |
+
"eval_accuracy": 0.9636678200692042,
|
| 1071 |
+
"eval_f1": 0.9635820582698403,
|
| 1072 |
+
"eval_loss": 0.24865780770778656,
|
| 1073 |
+
"eval_runtime": 0.9484,
|
| 1074 |
+
"eval_samples_per_second": 4266.086,
|
| 1075 |
+
"eval_steps_per_second": 67.481,
|
| 1076 |
+
"step": 14160
|
| 1077 |
+
}
|
| 1078 |
+
],
|
| 1079 |
+
"logging_steps": 100,
|
| 1080 |
+
"max_steps": 35400,
|
| 1081 |
+
"num_input_tokens_seen": 0,
|
| 1082 |
+
"num_train_epochs": 20,
|
| 1083 |
+
"save_steps": 500,
|
| 1084 |
+
"stateful_callbacks": {
|
| 1085 |
+
"EarlyStoppingCallback": {
|
| 1086 |
+
"args": {
|
| 1087 |
+
"early_stopping_patience": 3,
|
| 1088 |
+
"early_stopping_threshold": 0.0
|
| 1089 |
+
},
|
| 1090 |
+
"attributes": {
|
| 1091 |
+
"early_stopping_patience_counter": 0
|
| 1092 |
+
}
|
| 1093 |
+
},
|
| 1094 |
+
"TrainerControl": {
|
| 1095 |
+
"args": {
|
| 1096 |
+
"should_epoch_stop": false,
|
| 1097 |
+
"should_evaluate": false,
|
| 1098 |
+
"should_log": false,
|
| 1099 |
+
"should_save": true,
|
| 1100 |
+
"should_training_stop": false
|
| 1101 |
+
},
|
| 1102 |
+
"attributes": {}
|
| 1103 |
+
}
|
| 1104 |
+
},
|
| 1105 |
+
"total_flos": 1172336478105600.0,
|
| 1106 |
+
"train_batch_size": 16,
|
| 1107 |
+
"trial_name": null,
|
| 1108 |
+
"trial_params": null
|
| 1109 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a911aa07d9ca1f76618367adcad9ebf5a951be0ecd3ebd50dbcf89c173eadc77
|
| 3 |
+
size 5304
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|