Maxime Turlot
commited on
Commit
·
2e30d56
1
Parent(s):
bf39c9c
Add RedSecureBERT weights & tokenizer v1.0
Browse files- checkpoint-3600/config.json +27 -0
- checkpoint-3600/merges.txt +0 -0
- checkpoint-3600/model.safetensors +3 -0
- checkpoint-3600/optimizer.pt +3 -0
- checkpoint-3600/rng_state.pth +3 -0
- checkpoint-3600/scaler.pt +3 -0
- checkpoint-3600/scheduler.pt +3 -0
- checkpoint-3600/special_tokens_map.json +51 -0
- checkpoint-3600/tokenizer.json +0 -0
- checkpoint-3600/tokenizer_config.json +59 -0
- checkpoint-3600/trainer_state.json +439 -0
- checkpoint-3600/training_args.bin +3 -0
- checkpoint-3600/vocab.json +0 -0
- checkpoint-3804/config.json +27 -0
- checkpoint-3804/merges.txt +0 -0
- checkpoint-3804/model.safetensors +3 -0
- checkpoint-3804/optimizer.pt +3 -0
- checkpoint-3804/rng_state.pth +3 -0
- checkpoint-3804/scaler.pt +3 -0
- checkpoint-3804/scheduler.pt +3 -0
- checkpoint-3804/special_tokens_map.json +51 -0
- checkpoint-3804/tokenizer.json +0 -0
- checkpoint-3804/tokenizer_config.json +59 -0
- checkpoint-3804/trainer_state.json +453 -0
- checkpoint-3804/training_args.bin +3 -0
- checkpoint-3804/vocab.json +0 -0
- config.json +27 -0
- merges.txt +0 -0
- model.safetensors +3 -0
- special_tokens_map.json +51 -0
- tokenizer.json +0 -0
- tokenizer_config.json +59 -0
- training_args.bin +3 -0
- vocab.json +0 -0
checkpoint-3600/config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"RobertaForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"bos_token_id": 0,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"eos_token_id": 2,
|
| 9 |
+
"gradient_checkpointing": false,
|
| 10 |
+
"hidden_act": "gelu",
|
| 11 |
+
"hidden_dropout_prob": 0.1,
|
| 12 |
+
"hidden_size": 768,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 3072,
|
| 15 |
+
"layer_norm_eps": 1e-05,
|
| 16 |
+
"max_position_embeddings": 514,
|
| 17 |
+
"model_type": "roberta",
|
| 18 |
+
"num_attention_heads": 12,
|
| 19 |
+
"num_hidden_layers": 12,
|
| 20 |
+
"pad_token_id": 1,
|
| 21 |
+
"position_embedding_type": "absolute",
|
| 22 |
+
"torch_dtype": "float32",
|
| 23 |
+
"transformers_version": "4.50.3",
|
| 24 |
+
"type_vocab_size": 1,
|
| 25 |
+
"use_cache": true,
|
| 26 |
+
"vocab_size": 50265
|
| 27 |
+
}
|
checkpoint-3600/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-3600/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4bde84d20a445792ba3a143328eb646557707ffd76c8af96089c3e55bcf92b44
|
| 3 |
+
size 498612824
|
checkpoint-3600/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:93c8ca463bb4b02571797dd40657e6a96e2d8e417c5e85113620a29cfa8d61f4
|
| 3 |
+
size 997345530
|
checkpoint-3600/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1a1aa251bb255176db30dd4c1c3d83f585db66d30eaa13d0ce5faeab8760db28
|
| 3 |
+
size 14244
|
checkpoint-3600/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:21129cbeed301366e7d1fbac08af9534a54bee0e76cd5c5edced4c89537c6a18
|
| 3 |
+
size 988
|
checkpoint-3600/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6087704a70421504c6fbfaf61abcc4dd4445be298f97b1388cedc65c56f0bd84
|
| 3 |
+
size 1064
|
checkpoint-3600/special_tokens_map.json
ADDED
|
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": true,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"cls_token": {
|
| 10 |
+
"content": "<s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": true,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"eos_token": {
|
| 17 |
+
"content": "</s>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": true,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"mask_token": {
|
| 24 |
+
"content": "<mask>",
|
| 25 |
+
"lstrip": true,
|
| 26 |
+
"normalized": true,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
},
|
| 30 |
+
"pad_token": {
|
| 31 |
+
"content": "<pad>",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": true,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false
|
| 36 |
+
},
|
| 37 |
+
"sep_token": {
|
| 38 |
+
"content": "</s>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": true,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false
|
| 43 |
+
},
|
| 44 |
+
"unk_token": {
|
| 45 |
+
"content": "<unk>",
|
| 46 |
+
"lstrip": false,
|
| 47 |
+
"normalized": true,
|
| 48 |
+
"rstrip": false,
|
| 49 |
+
"single_word": false
|
| 50 |
+
}
|
| 51 |
+
}
|
checkpoint-3600/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-3600/tokenizer_config.json
ADDED
|
@@ -0,0 +1,59 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"0": {
|
| 5 |
+
"content": "<s>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": true,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"1": {
|
| 13 |
+
"content": "<pad>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": true,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"2": {
|
| 21 |
+
"content": "</s>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": true,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": true
|
| 27 |
+
},
|
| 28 |
+
"3": {
|
| 29 |
+
"content": "<unk>",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": true,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false,
|
| 34 |
+
"special": true
|
| 35 |
+
},
|
| 36 |
+
"50264": {
|
| 37 |
+
"content": "<mask>",
|
| 38 |
+
"lstrip": true,
|
| 39 |
+
"normalized": true,
|
| 40 |
+
"rstrip": false,
|
| 41 |
+
"single_word": false,
|
| 42 |
+
"special": true
|
| 43 |
+
}
|
| 44 |
+
},
|
| 45 |
+
"bos_token": "<s>",
|
| 46 |
+
"clean_up_tokenization_spaces": false,
|
| 47 |
+
"cls_token": "<s>",
|
| 48 |
+
"eos_token": "</s>",
|
| 49 |
+
"errors": "replace",
|
| 50 |
+
"extra_special_tokens": {},
|
| 51 |
+
"mask_token": "<mask>",
|
| 52 |
+
"max_len": 512,
|
| 53 |
+
"model_max_length": 512,
|
| 54 |
+
"pad_token": "<pad>",
|
| 55 |
+
"sep_token": "</s>",
|
| 56 |
+
"tokenizer_class": "RobertaTokenizer",
|
| 57 |
+
"trim_offsets": true,
|
| 58 |
+
"unk_token": "<unk>"
|
| 59 |
+
}
|
checkpoint-3600/trainer_state.json
ADDED
|
@@ -0,0 +1,439 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 3600,
|
| 3 |
+
"best_metric": 0.9848043970255416,
|
| 4 |
+
"best_model_checkpoint": "/workspace/AI/Trend_Primus-FineWeb_Filtering-pipeline/securebert_finetuned/offensive_vs_rest/checkpoint-3600",
|
| 5 |
+
"epoch": 2.8391167192429023,
|
| 6 |
+
"eval_steps": 300,
|
| 7 |
+
"global_step": 3600,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.07886435331230283,
|
| 14 |
+
"grad_norm": 1.846426010131836,
|
| 15 |
+
"learning_rate": 5.600000000000001e-06,
|
| 16 |
+
"loss": 0.2696,
|
| 17 |
+
"step": 100
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.15772870662460567,
|
| 21 |
+
"grad_norm": 5.872535228729248,
|
| 22 |
+
"learning_rate": 1.1314285714285715e-05,
|
| 23 |
+
"loss": 0.1452,
|
| 24 |
+
"step": 200
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.23659305993690852,
|
| 28 |
+
"grad_norm": 1.9362571239471436,
|
| 29 |
+
"learning_rate": 1.702857142857143e-05,
|
| 30 |
+
"loss": 0.0772,
|
| 31 |
+
"step": 300
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.23659305993690852,
|
| 35 |
+
"eval_f1": 0.7849643551523007,
|
| 36 |
+
"eval_f2": 0.8937426210153483,
|
| 37 |
+
"eval_loss": 0.04672536998987198,
|
| 38 |
+
"eval_precision": 0.6525862068965518,
|
| 39 |
+
"eval_recall": 0.9847154471544716,
|
| 40 |
+
"eval_runtime": 24.5242,
|
| 41 |
+
"eval_samples_per_second": 735.193,
|
| 42 |
+
"eval_steps_per_second": 11.499,
|
| 43 |
+
"step": 300
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 0.31545741324921134,
|
| 47 |
+
"grad_norm": 5.031215667724609,
|
| 48 |
+
"learning_rate": 1.9722061378112335e-05,
|
| 49 |
+
"loss": 0.0442,
|
| 50 |
+
"step": 400
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.3943217665615142,
|
| 54 |
+
"grad_norm": 1.0280592441558838,
|
| 55 |
+
"learning_rate": 1.9143022582513028e-05,
|
| 56 |
+
"loss": 0.0364,
|
| 57 |
+
"step": 500
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.47318611987381703,
|
| 61 |
+
"grad_norm": 1.5782877206802368,
|
| 62 |
+
"learning_rate": 1.8563983786913724e-05,
|
| 63 |
+
"loss": 0.0305,
|
| 64 |
+
"step": 600
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.47318611987381703,
|
| 68 |
+
"eval_f1": 0.8569032979318055,
|
| 69 |
+
"eval_f2": 0.9358403027898174,
|
| 70 |
+
"eval_loss": 0.02299814671278,
|
| 71 |
+
"eval_precision": 0.7512864493996569,
|
| 72 |
+
"eval_recall": 0.9970731707317073,
|
| 73 |
+
"eval_runtime": 24.8283,
|
| 74 |
+
"eval_samples_per_second": 726.188,
|
| 75 |
+
"eval_steps_per_second": 11.358,
|
| 76 |
+
"step": 600
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 0.5520504731861199,
|
| 80 |
+
"grad_norm": 1.143188714981079,
|
| 81 |
+
"learning_rate": 1.798494499131442e-05,
|
| 82 |
+
"loss": 0.0311,
|
| 83 |
+
"step": 700
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 0.6309148264984227,
|
| 87 |
+
"grad_norm": 1.4598668813705444,
|
| 88 |
+
"learning_rate": 1.7405906195715113e-05,
|
| 89 |
+
"loss": 0.0273,
|
| 90 |
+
"step": 800
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.7097791798107256,
|
| 94 |
+
"grad_norm": 0.9353739619255066,
|
| 95 |
+
"learning_rate": 1.682686740011581e-05,
|
| 96 |
+
"loss": 0.0232,
|
| 97 |
+
"step": 900
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.7097791798107256,
|
| 101 |
+
"eval_f1": 0.9401762250734271,
|
| 102 |
+
"eval_f2": 0.9688415955142092,
|
| 103 |
+
"eval_loss": 0.01698540337383747,
|
| 104 |
+
"eval_precision": 0.8959929286977019,
|
| 105 |
+
"eval_recall": 0.9889430894308943,
|
| 106 |
+
"eval_runtime": 24.5846,
|
| 107 |
+
"eval_samples_per_second": 733.386,
|
| 108 |
+
"eval_steps_per_second": 11.471,
|
| 109 |
+
"step": 900
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.7886435331230284,
|
| 113 |
+
"grad_norm": 1.172129511833191,
|
| 114 |
+
"learning_rate": 1.6247828604516505e-05,
|
| 115 |
+
"loss": 0.0232,
|
| 116 |
+
"step": 1000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.8675078864353313,
|
| 120 |
+
"grad_norm": 0.7822222113609314,
|
| 121 |
+
"learning_rate": 1.56687898089172e-05,
|
| 122 |
+
"loss": 0.0217,
|
| 123 |
+
"step": 1100
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.9463722397476341,
|
| 127 |
+
"grad_norm": 0.9808489680290222,
|
| 128 |
+
"learning_rate": 1.5089751013317892e-05,
|
| 129 |
+
"loss": 0.0167,
|
| 130 |
+
"step": 1200
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.9463722397476341,
|
| 134 |
+
"eval_f1": 0.9422098936662043,
|
| 135 |
+
"eval_f2": 0.9726995036273387,
|
| 136 |
+
"eval_loss": 0.013439147733151913,
|
| 137 |
+
"eval_precision": 0.8954305799648506,
|
| 138 |
+
"eval_recall": 0.9941463414634146,
|
| 139 |
+
"eval_runtime": 25.9737,
|
| 140 |
+
"eval_samples_per_second": 694.163,
|
| 141 |
+
"eval_steps_per_second": 10.857,
|
| 142 |
+
"step": 1200
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 1.025236593059937,
|
| 146 |
+
"grad_norm": 0.4348973035812378,
|
| 147 |
+
"learning_rate": 1.4510712217718588e-05,
|
| 148 |
+
"loss": 0.0162,
|
| 149 |
+
"step": 1300
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 1.1041009463722398,
|
| 153 |
+
"grad_norm": 0.4453680217266083,
|
| 154 |
+
"learning_rate": 1.3931673422119283e-05,
|
| 155 |
+
"loss": 0.0135,
|
| 156 |
+
"step": 1400
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 1.1829652996845426,
|
| 160 |
+
"grad_norm": 1.2444119453430176,
|
| 161 |
+
"learning_rate": 1.3352634626519977e-05,
|
| 162 |
+
"loss": 0.0093,
|
| 163 |
+
"step": 1500
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 1.1829652996845426,
|
| 167 |
+
"eval_f1": 0.9642065251821349,
|
| 168 |
+
"eval_f2": 0.9794710084304009,
|
| 169 |
+
"eval_loss": 0.014342821203172207,
|
| 170 |
+
"eval_precision": 0.939796233405372,
|
| 171 |
+
"eval_recall": 0.9899186991869918,
|
| 172 |
+
"eval_runtime": 24.7342,
|
| 173 |
+
"eval_samples_per_second": 728.949,
|
| 174 |
+
"eval_steps_per_second": 11.401,
|
| 175 |
+
"step": 1500
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 1.2618296529968454,
|
| 179 |
+
"grad_norm": 1.2138129472732544,
|
| 180 |
+
"learning_rate": 1.2773595830920673e-05,
|
| 181 |
+
"loss": 0.0106,
|
| 182 |
+
"step": 1600
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 1.3406940063091484,
|
| 186 |
+
"grad_norm": 3.329469680786133,
|
| 187 |
+
"learning_rate": 1.2194557035321368e-05,
|
| 188 |
+
"loss": 0.0101,
|
| 189 |
+
"step": 1700
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 1.4195583596214512,
|
| 193 |
+
"grad_norm": 0.7627914547920227,
|
| 194 |
+
"learning_rate": 1.1615518239722064e-05,
|
| 195 |
+
"loss": 0.0118,
|
| 196 |
+
"step": 1800
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 1.4195583596214512,
|
| 200 |
+
"eval_f1": 0.9513143568206563,
|
| 201 |
+
"eval_f2": 0.9767471572760955,
|
| 202 |
+
"eval_loss": 0.01234134566038847,
|
| 203 |
+
"eval_precision": 0.9117471675611211,
|
| 204 |
+
"eval_recall": 0.9944715447154472,
|
| 205 |
+
"eval_runtime": 25.1145,
|
| 206 |
+
"eval_samples_per_second": 717.911,
|
| 207 |
+
"eval_steps_per_second": 11.229,
|
| 208 |
+
"step": 1800
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 1.498422712933754,
|
| 212 |
+
"grad_norm": 0.9591709971427917,
|
| 213 |
+
"learning_rate": 1.1036479444122757e-05,
|
| 214 |
+
"loss": 0.0093,
|
| 215 |
+
"step": 1900
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 1.5772870662460567,
|
| 219 |
+
"grad_norm": 0.4569564759731293,
|
| 220 |
+
"learning_rate": 1.0457440648523451e-05,
|
| 221 |
+
"loss": 0.0094,
|
| 222 |
+
"step": 2000
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 1.6561514195583595,
|
| 226 |
+
"grad_norm": 0.7519212365150452,
|
| 227 |
+
"learning_rate": 9.88419224088014e-06,
|
| 228 |
+
"loss": 0.0094,
|
| 229 |
+
"step": 2100
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 1.6561514195583595,
|
| 233 |
+
"eval_f1": 0.9646464646464646,
|
| 234 |
+
"eval_f2": 0.9819420345736135,
|
| 235 |
+
"eval_loss": 0.012274333275854588,
|
| 236 |
+
"eval_precision": 0.9371358478994174,
|
| 237 |
+
"eval_recall": 0.9938211382113821,
|
| 238 |
+
"eval_runtime": 24.295,
|
| 239 |
+
"eval_samples_per_second": 742.127,
|
| 240 |
+
"eval_steps_per_second": 11.607,
|
| 241 |
+
"step": 2100
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.7350157728706623,
|
| 245 |
+
"grad_norm": 0.06854517012834549,
|
| 246 |
+
"learning_rate": 9.305153445280834e-06,
|
| 247 |
+
"loss": 0.0101,
|
| 248 |
+
"step": 2200
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.8138801261829653,
|
| 252 |
+
"grad_norm": 1.0062646865844727,
|
| 253 |
+
"learning_rate": 8.726114649681529e-06,
|
| 254 |
+
"loss": 0.0106,
|
| 255 |
+
"step": 2300
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 1.8927444794952681,
|
| 259 |
+
"grad_norm": 0.1466594785451889,
|
| 260 |
+
"learning_rate": 8.147075854082223e-06,
|
| 261 |
+
"loss": 0.0079,
|
| 262 |
+
"step": 2400
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 1.8927444794952681,
|
| 266 |
+
"eval_f1": 0.9619496855345911,
|
| 267 |
+
"eval_f2": 0.9813923644529997,
|
| 268 |
+
"eval_loss": 0.011407392099499702,
|
| 269 |
+
"eval_precision": 0.9312024353120244,
|
| 270 |
+
"eval_recall": 0.9947967479674796,
|
| 271 |
+
"eval_runtime": 26.3939,
|
| 272 |
+
"eval_samples_per_second": 683.112,
|
| 273 |
+
"eval_steps_per_second": 10.684,
|
| 274 |
+
"step": 2400
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 1.971608832807571,
|
| 278 |
+
"grad_norm": 0.26108694076538086,
|
| 279 |
+
"learning_rate": 7.568037058482919e-06,
|
| 280 |
+
"loss": 0.0094,
|
| 281 |
+
"step": 2500
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 2.050473186119874,
|
| 285 |
+
"grad_norm": 0.024676967412233353,
|
| 286 |
+
"learning_rate": 6.988998262883614e-06,
|
| 287 |
+
"loss": 0.0073,
|
| 288 |
+
"step": 2600
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 2.1293375394321767,
|
| 292 |
+
"grad_norm": 1.5645203590393066,
|
| 293 |
+
"learning_rate": 6.409959467284309e-06,
|
| 294 |
+
"loss": 0.0041,
|
| 295 |
+
"step": 2700
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 2.1293375394321767,
|
| 299 |
+
"eval_f1": 0.9681407513076558,
|
| 300 |
+
"eval_f2": 0.9830050212437235,
|
| 301 |
+
"eval_loss": 0.011503643356263638,
|
| 302 |
+
"eval_precision": 0.9443413729128015,
|
| 303 |
+
"eval_recall": 0.9931707317073171,
|
| 304 |
+
"eval_runtime": 25.3938,
|
| 305 |
+
"eval_samples_per_second": 710.014,
|
| 306 |
+
"eval_steps_per_second": 11.105,
|
| 307 |
+
"step": 2700
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 2.2082018927444795,
|
| 311 |
+
"grad_norm": 0.933417022228241,
|
| 312 |
+
"learning_rate": 5.830920671685003e-06,
|
| 313 |
+
"loss": 0.0045,
|
| 314 |
+
"step": 2800
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 2.2870662460567823,
|
| 318 |
+
"grad_norm": 0.7878792881965637,
|
| 319 |
+
"learning_rate": 5.251881876085698e-06,
|
| 320 |
+
"loss": 0.0037,
|
| 321 |
+
"step": 2900
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 2.365930599369085,
|
| 325 |
+
"grad_norm": 0.09505568444728851,
|
| 326 |
+
"learning_rate": 4.6728430804863925e-06,
|
| 327 |
+
"loss": 0.0035,
|
| 328 |
+
"step": 3000
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 2.365930599369085,
|
| 332 |
+
"eval_f1": 0.9727229223161589,
|
| 333 |
+
"eval_f2": 0.9839292629404931,
|
| 334 |
+
"eval_loss": 0.013037587516009808,
|
| 335 |
+
"eval_precision": 0.9546023794614903,
|
| 336 |
+
"eval_recall": 0.9915447154471545,
|
| 337 |
+
"eval_runtime": 26.4329,
|
| 338 |
+
"eval_samples_per_second": 682.105,
|
| 339 |
+
"eval_steps_per_second": 10.669,
|
| 340 |
+
"step": 3000
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 2.444794952681388,
|
| 344 |
+
"grad_norm": 0.08773530274629593,
|
| 345 |
+
"learning_rate": 4.093804284887088e-06,
|
| 346 |
+
"loss": 0.0029,
|
| 347 |
+
"step": 3100
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 2.5236593059936907,
|
| 351 |
+
"grad_norm": 0.060790352523326874,
|
| 352 |
+
"learning_rate": 3.5147654892877827e-06,
|
| 353 |
+
"loss": 0.0039,
|
| 354 |
+
"step": 3200
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 2.6025236593059935,
|
| 358 |
+
"grad_norm": 1.6132954359054565,
|
| 359 |
+
"learning_rate": 2.9357266936884776e-06,
|
| 360 |
+
"loss": 0.0043,
|
| 361 |
+
"step": 3300
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 2.6025236593059935,
|
| 365 |
+
"eval_f1": 0.975609756097561,
|
| 366 |
+
"eval_f2": 0.9833732289577538,
|
| 367 |
+
"eval_loss": 0.014466837979853153,
|
| 368 |
+
"eval_precision": 0.9629394995248653,
|
| 369 |
+
"eval_recall": 0.9886178861788618,
|
| 370 |
+
"eval_runtime": 26.7793,
|
| 371 |
+
"eval_samples_per_second": 673.282,
|
| 372 |
+
"eval_steps_per_second": 10.531,
|
| 373 |
+
"step": 3300
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 2.6813880126182967,
|
| 377 |
+
"grad_norm": 0.24806837737560272,
|
| 378 |
+
"learning_rate": 2.356687898089172e-06,
|
| 379 |
+
"loss": 0.0047,
|
| 380 |
+
"step": 3400
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 2.7602523659305995,
|
| 384 |
+
"grad_norm": 3.355231523513794,
|
| 385 |
+
"learning_rate": 1.777649102489867e-06,
|
| 386 |
+
"loss": 0.0031,
|
| 387 |
+
"step": 3500
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 2.8391167192429023,
|
| 391 |
+
"grad_norm": 0.44861266016960144,
|
| 392 |
+
"learning_rate": 1.1986103068905617e-06,
|
| 393 |
+
"loss": 0.004,
|
| 394 |
+
"step": 3600
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 2.8391167192429023,
|
| 398 |
+
"eval_f1": 0.9762820512820513,
|
| 399 |
+
"eval_f2": 0.9848043970255416,
|
| 400 |
+
"eval_loss": 0.01391169149428606,
|
| 401 |
+
"eval_precision": 0.9624012638230648,
|
| 402 |
+
"eval_recall": 0.9905691056910569,
|
| 403 |
+
"eval_runtime": 26.114,
|
| 404 |
+
"eval_samples_per_second": 690.435,
|
| 405 |
+
"eval_steps_per_second": 10.799,
|
| 406 |
+
"step": 3600
|
| 407 |
+
}
|
| 408 |
+
],
|
| 409 |
+
"logging_steps": 100,
|
| 410 |
+
"max_steps": 3804,
|
| 411 |
+
"num_input_tokens_seen": 0,
|
| 412 |
+
"num_train_epochs": 3,
|
| 413 |
+
"save_steps": 300,
|
| 414 |
+
"stateful_callbacks": {
|
| 415 |
+
"EarlyStoppingCallback": {
|
| 416 |
+
"args": {
|
| 417 |
+
"early_stopping_patience": 2,
|
| 418 |
+
"early_stopping_threshold": 0.0
|
| 419 |
+
},
|
| 420 |
+
"attributes": {
|
| 421 |
+
"early_stopping_patience_counter": 0
|
| 422 |
+
}
|
| 423 |
+
},
|
| 424 |
+
"TrainerControl": {
|
| 425 |
+
"args": {
|
| 426 |
+
"should_epoch_stop": false,
|
| 427 |
+
"should_evaluate": false,
|
| 428 |
+
"should_log": false,
|
| 429 |
+
"should_save": true,
|
| 430 |
+
"should_training_stop": false
|
| 431 |
+
},
|
| 432 |
+
"attributes": {}
|
| 433 |
+
}
|
| 434 |
+
},
|
| 435 |
+
"total_flos": 1.2122157786968064e+17,
|
| 436 |
+
"train_batch_size": 64,
|
| 437 |
+
"trial_name": null,
|
| 438 |
+
"trial_params": null
|
| 439 |
+
}
|
checkpoint-3600/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cd899628771df754e988ea5f7948893d128805d0ba8d41adf70edde62e2a37f1
|
| 3 |
+
size 5496
|
checkpoint-3600/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-3804/config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"RobertaForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"bos_token_id": 0,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"eos_token_id": 2,
|
| 9 |
+
"gradient_checkpointing": false,
|
| 10 |
+
"hidden_act": "gelu",
|
| 11 |
+
"hidden_dropout_prob": 0.1,
|
| 12 |
+
"hidden_size": 768,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 3072,
|
| 15 |
+
"layer_norm_eps": 1e-05,
|
| 16 |
+
"max_position_embeddings": 514,
|
| 17 |
+
"model_type": "roberta",
|
| 18 |
+
"num_attention_heads": 12,
|
| 19 |
+
"num_hidden_layers": 12,
|
| 20 |
+
"pad_token_id": 1,
|
| 21 |
+
"position_embedding_type": "absolute",
|
| 22 |
+
"torch_dtype": "float32",
|
| 23 |
+
"transformers_version": "4.50.3",
|
| 24 |
+
"type_vocab_size": 1,
|
| 25 |
+
"use_cache": true,
|
| 26 |
+
"vocab_size": 50265
|
| 27 |
+
}
|
checkpoint-3804/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-3804/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:71172f6e2b5f78b654120da8b3b00a70dcd3697c4375cfee025ebf1f3cc0cf93
|
| 3 |
+
size 498612824
|
checkpoint-3804/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:61fae529836705567644118b09f9966479a0a067dd564171dffcbbd1d60f0624
|
| 3 |
+
size 997345530
|
checkpoint-3804/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0221df9a77e7e8d8a3eae8354c62cbbede66a8ce42095aafba80c2afab538970
|
| 3 |
+
size 14244
|
checkpoint-3804/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:678047e33b981c0fce5eb0814029e08bb444a7c31df70503532fa7964ac9d661
|
| 3 |
+
size 988
|
checkpoint-3804/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d9502a8b983bae7ca729013ff02a191e5739e07e4f66f9b75110407439647582
|
| 3 |
+
size 1064
|
checkpoint-3804/special_tokens_map.json
ADDED
|
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": true,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"cls_token": {
|
| 10 |
+
"content": "<s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": true,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"eos_token": {
|
| 17 |
+
"content": "</s>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": true,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"mask_token": {
|
| 24 |
+
"content": "<mask>",
|
| 25 |
+
"lstrip": true,
|
| 26 |
+
"normalized": true,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
},
|
| 30 |
+
"pad_token": {
|
| 31 |
+
"content": "<pad>",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": true,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false
|
| 36 |
+
},
|
| 37 |
+
"sep_token": {
|
| 38 |
+
"content": "</s>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": true,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false
|
| 43 |
+
},
|
| 44 |
+
"unk_token": {
|
| 45 |
+
"content": "<unk>",
|
| 46 |
+
"lstrip": false,
|
| 47 |
+
"normalized": true,
|
| 48 |
+
"rstrip": false,
|
| 49 |
+
"single_word": false
|
| 50 |
+
}
|
| 51 |
+
}
|
checkpoint-3804/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-3804/tokenizer_config.json
ADDED
|
@@ -0,0 +1,59 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"0": {
|
| 5 |
+
"content": "<s>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": true,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"1": {
|
| 13 |
+
"content": "<pad>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": true,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"2": {
|
| 21 |
+
"content": "</s>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": true,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": true
|
| 27 |
+
},
|
| 28 |
+
"3": {
|
| 29 |
+
"content": "<unk>",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": true,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false,
|
| 34 |
+
"special": true
|
| 35 |
+
},
|
| 36 |
+
"50264": {
|
| 37 |
+
"content": "<mask>",
|
| 38 |
+
"lstrip": true,
|
| 39 |
+
"normalized": true,
|
| 40 |
+
"rstrip": false,
|
| 41 |
+
"single_word": false,
|
| 42 |
+
"special": true
|
| 43 |
+
}
|
| 44 |
+
},
|
| 45 |
+
"bos_token": "<s>",
|
| 46 |
+
"clean_up_tokenization_spaces": false,
|
| 47 |
+
"cls_token": "<s>",
|
| 48 |
+
"eos_token": "</s>",
|
| 49 |
+
"errors": "replace",
|
| 50 |
+
"extra_special_tokens": {},
|
| 51 |
+
"mask_token": "<mask>",
|
| 52 |
+
"max_len": 512,
|
| 53 |
+
"model_max_length": 512,
|
| 54 |
+
"pad_token": "<pad>",
|
| 55 |
+
"sep_token": "</s>",
|
| 56 |
+
"tokenizer_class": "RobertaTokenizer",
|
| 57 |
+
"trim_offsets": true,
|
| 58 |
+
"unk_token": "<unk>"
|
| 59 |
+
}
|
checkpoint-3804/trainer_state.json
ADDED
|
@@ -0,0 +1,453 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 3600,
|
| 3 |
+
"best_metric": 0.9848043970255416,
|
| 4 |
+
"best_model_checkpoint": "/workspace/AI/Trend_Primus-FineWeb_Filtering-pipeline/securebert_finetuned/offensive_vs_rest/checkpoint-3600",
|
| 5 |
+
"epoch": 3.0,
|
| 6 |
+
"eval_steps": 300,
|
| 7 |
+
"global_step": 3804,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.07886435331230283,
|
| 14 |
+
"grad_norm": 1.846426010131836,
|
| 15 |
+
"learning_rate": 5.600000000000001e-06,
|
| 16 |
+
"loss": 0.2696,
|
| 17 |
+
"step": 100
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.15772870662460567,
|
| 21 |
+
"grad_norm": 5.872535228729248,
|
| 22 |
+
"learning_rate": 1.1314285714285715e-05,
|
| 23 |
+
"loss": 0.1452,
|
| 24 |
+
"step": 200
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.23659305993690852,
|
| 28 |
+
"grad_norm": 1.9362571239471436,
|
| 29 |
+
"learning_rate": 1.702857142857143e-05,
|
| 30 |
+
"loss": 0.0772,
|
| 31 |
+
"step": 300
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.23659305993690852,
|
| 35 |
+
"eval_f1": 0.7849643551523007,
|
| 36 |
+
"eval_f2": 0.8937426210153483,
|
| 37 |
+
"eval_loss": 0.04672536998987198,
|
| 38 |
+
"eval_precision": 0.6525862068965518,
|
| 39 |
+
"eval_recall": 0.9847154471544716,
|
| 40 |
+
"eval_runtime": 24.5242,
|
| 41 |
+
"eval_samples_per_second": 735.193,
|
| 42 |
+
"eval_steps_per_second": 11.499,
|
| 43 |
+
"step": 300
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 0.31545741324921134,
|
| 47 |
+
"grad_norm": 5.031215667724609,
|
| 48 |
+
"learning_rate": 1.9722061378112335e-05,
|
| 49 |
+
"loss": 0.0442,
|
| 50 |
+
"step": 400
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.3943217665615142,
|
| 54 |
+
"grad_norm": 1.0280592441558838,
|
| 55 |
+
"learning_rate": 1.9143022582513028e-05,
|
| 56 |
+
"loss": 0.0364,
|
| 57 |
+
"step": 500
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.47318611987381703,
|
| 61 |
+
"grad_norm": 1.5782877206802368,
|
| 62 |
+
"learning_rate": 1.8563983786913724e-05,
|
| 63 |
+
"loss": 0.0305,
|
| 64 |
+
"step": 600
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.47318611987381703,
|
| 68 |
+
"eval_f1": 0.8569032979318055,
|
| 69 |
+
"eval_f2": 0.9358403027898174,
|
| 70 |
+
"eval_loss": 0.02299814671278,
|
| 71 |
+
"eval_precision": 0.7512864493996569,
|
| 72 |
+
"eval_recall": 0.9970731707317073,
|
| 73 |
+
"eval_runtime": 24.8283,
|
| 74 |
+
"eval_samples_per_second": 726.188,
|
| 75 |
+
"eval_steps_per_second": 11.358,
|
| 76 |
+
"step": 600
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 0.5520504731861199,
|
| 80 |
+
"grad_norm": 1.143188714981079,
|
| 81 |
+
"learning_rate": 1.798494499131442e-05,
|
| 82 |
+
"loss": 0.0311,
|
| 83 |
+
"step": 700
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 0.6309148264984227,
|
| 87 |
+
"grad_norm": 1.4598668813705444,
|
| 88 |
+
"learning_rate": 1.7405906195715113e-05,
|
| 89 |
+
"loss": 0.0273,
|
| 90 |
+
"step": 800
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.7097791798107256,
|
| 94 |
+
"grad_norm": 0.9353739619255066,
|
| 95 |
+
"learning_rate": 1.682686740011581e-05,
|
| 96 |
+
"loss": 0.0232,
|
| 97 |
+
"step": 900
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.7097791798107256,
|
| 101 |
+
"eval_f1": 0.9401762250734271,
|
| 102 |
+
"eval_f2": 0.9688415955142092,
|
| 103 |
+
"eval_loss": 0.01698540337383747,
|
| 104 |
+
"eval_precision": 0.8959929286977019,
|
| 105 |
+
"eval_recall": 0.9889430894308943,
|
| 106 |
+
"eval_runtime": 24.5846,
|
| 107 |
+
"eval_samples_per_second": 733.386,
|
| 108 |
+
"eval_steps_per_second": 11.471,
|
| 109 |
+
"step": 900
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.7886435331230284,
|
| 113 |
+
"grad_norm": 1.172129511833191,
|
| 114 |
+
"learning_rate": 1.6247828604516505e-05,
|
| 115 |
+
"loss": 0.0232,
|
| 116 |
+
"step": 1000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.8675078864353313,
|
| 120 |
+
"grad_norm": 0.7822222113609314,
|
| 121 |
+
"learning_rate": 1.56687898089172e-05,
|
| 122 |
+
"loss": 0.0217,
|
| 123 |
+
"step": 1100
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.9463722397476341,
|
| 127 |
+
"grad_norm": 0.9808489680290222,
|
| 128 |
+
"learning_rate": 1.5089751013317892e-05,
|
| 129 |
+
"loss": 0.0167,
|
| 130 |
+
"step": 1200
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.9463722397476341,
|
| 134 |
+
"eval_f1": 0.9422098936662043,
|
| 135 |
+
"eval_f2": 0.9726995036273387,
|
| 136 |
+
"eval_loss": 0.013439147733151913,
|
| 137 |
+
"eval_precision": 0.8954305799648506,
|
| 138 |
+
"eval_recall": 0.9941463414634146,
|
| 139 |
+
"eval_runtime": 25.9737,
|
| 140 |
+
"eval_samples_per_second": 694.163,
|
| 141 |
+
"eval_steps_per_second": 10.857,
|
| 142 |
+
"step": 1200
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 1.025236593059937,
|
| 146 |
+
"grad_norm": 0.4348973035812378,
|
| 147 |
+
"learning_rate": 1.4510712217718588e-05,
|
| 148 |
+
"loss": 0.0162,
|
| 149 |
+
"step": 1300
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 1.1041009463722398,
|
| 153 |
+
"grad_norm": 0.4453680217266083,
|
| 154 |
+
"learning_rate": 1.3931673422119283e-05,
|
| 155 |
+
"loss": 0.0135,
|
| 156 |
+
"step": 1400
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 1.1829652996845426,
|
| 160 |
+
"grad_norm": 1.2444119453430176,
|
| 161 |
+
"learning_rate": 1.3352634626519977e-05,
|
| 162 |
+
"loss": 0.0093,
|
| 163 |
+
"step": 1500
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 1.1829652996845426,
|
| 167 |
+
"eval_f1": 0.9642065251821349,
|
| 168 |
+
"eval_f2": 0.9794710084304009,
|
| 169 |
+
"eval_loss": 0.014342821203172207,
|
| 170 |
+
"eval_precision": 0.939796233405372,
|
| 171 |
+
"eval_recall": 0.9899186991869918,
|
| 172 |
+
"eval_runtime": 24.7342,
|
| 173 |
+
"eval_samples_per_second": 728.949,
|
| 174 |
+
"eval_steps_per_second": 11.401,
|
| 175 |
+
"step": 1500
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 1.2618296529968454,
|
| 179 |
+
"grad_norm": 1.2138129472732544,
|
| 180 |
+
"learning_rate": 1.2773595830920673e-05,
|
| 181 |
+
"loss": 0.0106,
|
| 182 |
+
"step": 1600
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 1.3406940063091484,
|
| 186 |
+
"grad_norm": 3.329469680786133,
|
| 187 |
+
"learning_rate": 1.2194557035321368e-05,
|
| 188 |
+
"loss": 0.0101,
|
| 189 |
+
"step": 1700
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 1.4195583596214512,
|
| 193 |
+
"grad_norm": 0.7627914547920227,
|
| 194 |
+
"learning_rate": 1.1615518239722064e-05,
|
| 195 |
+
"loss": 0.0118,
|
| 196 |
+
"step": 1800
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 1.4195583596214512,
|
| 200 |
+
"eval_f1": 0.9513143568206563,
|
| 201 |
+
"eval_f2": 0.9767471572760955,
|
| 202 |
+
"eval_loss": 0.01234134566038847,
|
| 203 |
+
"eval_precision": 0.9117471675611211,
|
| 204 |
+
"eval_recall": 0.9944715447154472,
|
| 205 |
+
"eval_runtime": 25.1145,
|
| 206 |
+
"eval_samples_per_second": 717.911,
|
| 207 |
+
"eval_steps_per_second": 11.229,
|
| 208 |
+
"step": 1800
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 1.498422712933754,
|
| 212 |
+
"grad_norm": 0.9591709971427917,
|
| 213 |
+
"learning_rate": 1.1036479444122757e-05,
|
| 214 |
+
"loss": 0.0093,
|
| 215 |
+
"step": 1900
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 1.5772870662460567,
|
| 219 |
+
"grad_norm": 0.4569564759731293,
|
| 220 |
+
"learning_rate": 1.0457440648523451e-05,
|
| 221 |
+
"loss": 0.0094,
|
| 222 |
+
"step": 2000
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 1.6561514195583595,
|
| 226 |
+
"grad_norm": 0.7519212365150452,
|
| 227 |
+
"learning_rate": 9.88419224088014e-06,
|
| 228 |
+
"loss": 0.0094,
|
| 229 |
+
"step": 2100
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 1.6561514195583595,
|
| 233 |
+
"eval_f1": 0.9646464646464646,
|
| 234 |
+
"eval_f2": 0.9819420345736135,
|
| 235 |
+
"eval_loss": 0.012274333275854588,
|
| 236 |
+
"eval_precision": 0.9371358478994174,
|
| 237 |
+
"eval_recall": 0.9938211382113821,
|
| 238 |
+
"eval_runtime": 24.295,
|
| 239 |
+
"eval_samples_per_second": 742.127,
|
| 240 |
+
"eval_steps_per_second": 11.607,
|
| 241 |
+
"step": 2100
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.7350157728706623,
|
| 245 |
+
"grad_norm": 0.06854517012834549,
|
| 246 |
+
"learning_rate": 9.305153445280834e-06,
|
| 247 |
+
"loss": 0.0101,
|
| 248 |
+
"step": 2200
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.8138801261829653,
|
| 252 |
+
"grad_norm": 1.0062646865844727,
|
| 253 |
+
"learning_rate": 8.726114649681529e-06,
|
| 254 |
+
"loss": 0.0106,
|
| 255 |
+
"step": 2300
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 1.8927444794952681,
|
| 259 |
+
"grad_norm": 0.1466594785451889,
|
| 260 |
+
"learning_rate": 8.147075854082223e-06,
|
| 261 |
+
"loss": 0.0079,
|
| 262 |
+
"step": 2400
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 1.8927444794952681,
|
| 266 |
+
"eval_f1": 0.9619496855345911,
|
| 267 |
+
"eval_f2": 0.9813923644529997,
|
| 268 |
+
"eval_loss": 0.011407392099499702,
|
| 269 |
+
"eval_precision": 0.9312024353120244,
|
| 270 |
+
"eval_recall": 0.9947967479674796,
|
| 271 |
+
"eval_runtime": 26.3939,
|
| 272 |
+
"eval_samples_per_second": 683.112,
|
| 273 |
+
"eval_steps_per_second": 10.684,
|
| 274 |
+
"step": 2400
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 1.971608832807571,
|
| 278 |
+
"grad_norm": 0.26108694076538086,
|
| 279 |
+
"learning_rate": 7.568037058482919e-06,
|
| 280 |
+
"loss": 0.0094,
|
| 281 |
+
"step": 2500
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 2.050473186119874,
|
| 285 |
+
"grad_norm": 0.024676967412233353,
|
| 286 |
+
"learning_rate": 6.988998262883614e-06,
|
| 287 |
+
"loss": 0.0073,
|
| 288 |
+
"step": 2600
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 2.1293375394321767,
|
| 292 |
+
"grad_norm": 1.5645203590393066,
|
| 293 |
+
"learning_rate": 6.409959467284309e-06,
|
| 294 |
+
"loss": 0.0041,
|
| 295 |
+
"step": 2700
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 2.1293375394321767,
|
| 299 |
+
"eval_f1": 0.9681407513076558,
|
| 300 |
+
"eval_f2": 0.9830050212437235,
|
| 301 |
+
"eval_loss": 0.011503643356263638,
|
| 302 |
+
"eval_precision": 0.9443413729128015,
|
| 303 |
+
"eval_recall": 0.9931707317073171,
|
| 304 |
+
"eval_runtime": 25.3938,
|
| 305 |
+
"eval_samples_per_second": 710.014,
|
| 306 |
+
"eval_steps_per_second": 11.105,
|
| 307 |
+
"step": 2700
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 2.2082018927444795,
|
| 311 |
+
"grad_norm": 0.933417022228241,
|
| 312 |
+
"learning_rate": 5.830920671685003e-06,
|
| 313 |
+
"loss": 0.0045,
|
| 314 |
+
"step": 2800
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 2.2870662460567823,
|
| 318 |
+
"grad_norm": 0.7878792881965637,
|
| 319 |
+
"learning_rate": 5.251881876085698e-06,
|
| 320 |
+
"loss": 0.0037,
|
| 321 |
+
"step": 2900
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 2.365930599369085,
|
| 325 |
+
"grad_norm": 0.09505568444728851,
|
| 326 |
+
"learning_rate": 4.6728430804863925e-06,
|
| 327 |
+
"loss": 0.0035,
|
| 328 |
+
"step": 3000
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 2.365930599369085,
|
| 332 |
+
"eval_f1": 0.9727229223161589,
|
| 333 |
+
"eval_f2": 0.9839292629404931,
|
| 334 |
+
"eval_loss": 0.013037587516009808,
|
| 335 |
+
"eval_precision": 0.9546023794614903,
|
| 336 |
+
"eval_recall": 0.9915447154471545,
|
| 337 |
+
"eval_runtime": 26.4329,
|
| 338 |
+
"eval_samples_per_second": 682.105,
|
| 339 |
+
"eval_steps_per_second": 10.669,
|
| 340 |
+
"step": 3000
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 2.444794952681388,
|
| 344 |
+
"grad_norm": 0.08773530274629593,
|
| 345 |
+
"learning_rate": 4.093804284887088e-06,
|
| 346 |
+
"loss": 0.0029,
|
| 347 |
+
"step": 3100
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 2.5236593059936907,
|
| 351 |
+
"grad_norm": 0.060790352523326874,
|
| 352 |
+
"learning_rate": 3.5147654892877827e-06,
|
| 353 |
+
"loss": 0.0039,
|
| 354 |
+
"step": 3200
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 2.6025236593059935,
|
| 358 |
+
"grad_norm": 1.6132954359054565,
|
| 359 |
+
"learning_rate": 2.9357266936884776e-06,
|
| 360 |
+
"loss": 0.0043,
|
| 361 |
+
"step": 3300
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 2.6025236593059935,
|
| 365 |
+
"eval_f1": 0.975609756097561,
|
| 366 |
+
"eval_f2": 0.9833732289577538,
|
| 367 |
+
"eval_loss": 0.014466837979853153,
|
| 368 |
+
"eval_precision": 0.9629394995248653,
|
| 369 |
+
"eval_recall": 0.9886178861788618,
|
| 370 |
+
"eval_runtime": 26.7793,
|
| 371 |
+
"eval_samples_per_second": 673.282,
|
| 372 |
+
"eval_steps_per_second": 10.531,
|
| 373 |
+
"step": 3300
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 2.6813880126182967,
|
| 377 |
+
"grad_norm": 0.24806837737560272,
|
| 378 |
+
"learning_rate": 2.356687898089172e-06,
|
| 379 |
+
"loss": 0.0047,
|
| 380 |
+
"step": 3400
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 2.7602523659305995,
|
| 384 |
+
"grad_norm": 3.355231523513794,
|
| 385 |
+
"learning_rate": 1.777649102489867e-06,
|
| 386 |
+
"loss": 0.0031,
|
| 387 |
+
"step": 3500
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 2.8391167192429023,
|
| 391 |
+
"grad_norm": 0.44861266016960144,
|
| 392 |
+
"learning_rate": 1.1986103068905617e-06,
|
| 393 |
+
"loss": 0.004,
|
| 394 |
+
"step": 3600
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 2.8391167192429023,
|
| 398 |
+
"eval_f1": 0.9762820512820513,
|
| 399 |
+
"eval_f2": 0.9848043970255416,
|
| 400 |
+
"eval_loss": 0.01391169149428606,
|
| 401 |
+
"eval_precision": 0.9624012638230648,
|
| 402 |
+
"eval_recall": 0.9905691056910569,
|
| 403 |
+
"eval_runtime": 26.114,
|
| 404 |
+
"eval_samples_per_second": 690.435,
|
| 405 |
+
"eval_steps_per_second": 10.799,
|
| 406 |
+
"step": 3600
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 2.917981072555205,
|
| 410 |
+
"grad_norm": 2.625840663909912,
|
| 411 |
+
"learning_rate": 6.195715112912566e-07,
|
| 412 |
+
"loss": 0.0032,
|
| 413 |
+
"step": 3700
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 2.996845425867508,
|
| 417 |
+
"grad_norm": 0.7660179138183594,
|
| 418 |
+
"learning_rate": 4.053271569195137e-08,
|
| 419 |
+
"loss": 0.003,
|
| 420 |
+
"step": 3800
|
| 421 |
+
}
|
| 422 |
+
],
|
| 423 |
+
"logging_steps": 100,
|
| 424 |
+
"max_steps": 3804,
|
| 425 |
+
"num_input_tokens_seen": 0,
|
| 426 |
+
"num_train_epochs": 3,
|
| 427 |
+
"save_steps": 300,
|
| 428 |
+
"stateful_callbacks": {
|
| 429 |
+
"EarlyStoppingCallback": {
|
| 430 |
+
"args": {
|
| 431 |
+
"early_stopping_patience": 2,
|
| 432 |
+
"early_stopping_threshold": 0.0
|
| 433 |
+
},
|
| 434 |
+
"attributes": {
|
| 435 |
+
"early_stopping_patience_counter": 0
|
| 436 |
+
}
|
| 437 |
+
},
|
| 438 |
+
"TrainerControl": {
|
| 439 |
+
"args": {
|
| 440 |
+
"should_epoch_stop": false,
|
| 441 |
+
"should_evaluate": false,
|
| 442 |
+
"should_log": false,
|
| 443 |
+
"should_save": true,
|
| 444 |
+
"should_training_stop": true
|
| 445 |
+
},
|
| 446 |
+
"attributes": {}
|
| 447 |
+
}
|
| 448 |
+
},
|
| 449 |
+
"total_flos": 1.2808193552713728e+17,
|
| 450 |
+
"train_batch_size": 64,
|
| 451 |
+
"trial_name": null,
|
| 452 |
+
"trial_params": null
|
| 453 |
+
}
|
checkpoint-3804/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cd899628771df754e988ea5f7948893d128805d0ba8d41adf70edde62e2a37f1
|
| 3 |
+
size 5496
|
checkpoint-3804/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"RobertaForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"bos_token_id": 0,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"eos_token_id": 2,
|
| 9 |
+
"gradient_checkpointing": false,
|
| 10 |
+
"hidden_act": "gelu",
|
| 11 |
+
"hidden_dropout_prob": 0.1,
|
| 12 |
+
"hidden_size": 768,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 3072,
|
| 15 |
+
"layer_norm_eps": 1e-05,
|
| 16 |
+
"max_position_embeddings": 514,
|
| 17 |
+
"model_type": "roberta",
|
| 18 |
+
"num_attention_heads": 12,
|
| 19 |
+
"num_hidden_layers": 12,
|
| 20 |
+
"pad_token_id": 1,
|
| 21 |
+
"position_embedding_type": "absolute",
|
| 22 |
+
"torch_dtype": "float32",
|
| 23 |
+
"transformers_version": "4.50.3",
|
| 24 |
+
"type_vocab_size": 1,
|
| 25 |
+
"use_cache": true,
|
| 26 |
+
"vocab_size": 50265
|
| 27 |
+
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4bde84d20a445792ba3a143328eb646557707ffd76c8af96089c3e55bcf92b44
|
| 3 |
+
size 498612824
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": true,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"cls_token": {
|
| 10 |
+
"content": "<s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": true,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"eos_token": {
|
| 17 |
+
"content": "</s>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": true,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"mask_token": {
|
| 24 |
+
"content": "<mask>",
|
| 25 |
+
"lstrip": true,
|
| 26 |
+
"normalized": true,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
},
|
| 30 |
+
"pad_token": {
|
| 31 |
+
"content": "<pad>",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": true,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false
|
| 36 |
+
},
|
| 37 |
+
"sep_token": {
|
| 38 |
+
"content": "</s>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": true,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false
|
| 43 |
+
},
|
| 44 |
+
"unk_token": {
|
| 45 |
+
"content": "<unk>",
|
| 46 |
+
"lstrip": false,
|
| 47 |
+
"normalized": true,
|
| 48 |
+
"rstrip": false,
|
| 49 |
+
"single_word": false
|
| 50 |
+
}
|
| 51 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,59 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"0": {
|
| 5 |
+
"content": "<s>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": true,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"1": {
|
| 13 |
+
"content": "<pad>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": true,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"2": {
|
| 21 |
+
"content": "</s>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": true,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": true
|
| 27 |
+
},
|
| 28 |
+
"3": {
|
| 29 |
+
"content": "<unk>",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": true,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false,
|
| 34 |
+
"special": true
|
| 35 |
+
},
|
| 36 |
+
"50264": {
|
| 37 |
+
"content": "<mask>",
|
| 38 |
+
"lstrip": true,
|
| 39 |
+
"normalized": true,
|
| 40 |
+
"rstrip": false,
|
| 41 |
+
"single_word": false,
|
| 42 |
+
"special": true
|
| 43 |
+
}
|
| 44 |
+
},
|
| 45 |
+
"bos_token": "<s>",
|
| 46 |
+
"clean_up_tokenization_spaces": false,
|
| 47 |
+
"cls_token": "<s>",
|
| 48 |
+
"eos_token": "</s>",
|
| 49 |
+
"errors": "replace",
|
| 50 |
+
"extra_special_tokens": {},
|
| 51 |
+
"mask_token": "<mask>",
|
| 52 |
+
"max_len": 512,
|
| 53 |
+
"model_max_length": 512,
|
| 54 |
+
"pad_token": "<pad>",
|
| 55 |
+
"sep_token": "</s>",
|
| 56 |
+
"tokenizer_class": "RobertaTokenizer",
|
| 57 |
+
"trim_offsets": true,
|
| 58 |
+
"unk_token": "<unk>"
|
| 59 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cd899628771df754e988ea5f7948893d128805d0ba8d41adf70edde62e2a37f1
|
| 3 |
+
size 5496
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|