Upload 12 files
Browse files- added_tokens.json +6 -0
- all_results.json +16 -0
- config.json +35 -0
- eval_results.json +10 -0
- model.safetensors +3 -0
- special_tokens_map.json +43 -0
- tokenizer.json +0 -0
- tokenizer_config.json +103 -0
- train_results.json +10 -0
- trainer_state.json +744 -0
- training_args.bin +3 -0
- vocab.txt +0 -0
added_tokens.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<|...|>": 30525,
|
| 3 |
+
"<|assistant|>": 30524,
|
| 4 |
+
"<|system|>": 30522,
|
| 5 |
+
"<|user|>": 30523
|
| 6 |
+
}
|
all_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_accuracy": 0.919369715603382,
|
| 4 |
+
"eval_loss": 0.19649724662303925,
|
| 5 |
+
"eval_runtime": 5.3689,
|
| 6 |
+
"eval_samples": 13010,
|
| 7 |
+
"eval_samples_per_second": 2423.215,
|
| 8 |
+
"eval_steps_per_second": 303.042,
|
| 9 |
+
"num_input_tokens_seen": 33305600,
|
| 10 |
+
"total_flos": 2157435918643200.0,
|
| 11 |
+
"train_loss": 0.1371641572881533,
|
| 12 |
+
"train_runtime": 362.4532,
|
| 13 |
+
"train_samples": 52040,
|
| 14 |
+
"train_samples_per_second": 717.886,
|
| 15 |
+
"train_steps_per_second": 89.736
|
| 16 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"BertForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"classifier_dropout": null,
|
| 7 |
+
"dtype": "float32",
|
| 8 |
+
"finetuning_task": "text-classification",
|
| 9 |
+
"gradient_checkpointing": false,
|
| 10 |
+
"hidden_act": "gelu",
|
| 11 |
+
"hidden_dropout_prob": 0.1,
|
| 12 |
+
"hidden_size": 384,
|
| 13 |
+
"id2label": {
|
| 14 |
+
"0": "Non-refusal",
|
| 15 |
+
"1": "Refusal"
|
| 16 |
+
},
|
| 17 |
+
"initializer_range": 0.02,
|
| 18 |
+
"intermediate_size": 1536,
|
| 19 |
+
"label2id": {
|
| 20 |
+
"Non-refusal": 0,
|
| 21 |
+
"Refusal": 1
|
| 22 |
+
},
|
| 23 |
+
"layer_norm_eps": 1e-12,
|
| 24 |
+
"max_position_embeddings": 512,
|
| 25 |
+
"model_type": "bert",
|
| 26 |
+
"num_attention_heads": 12,
|
| 27 |
+
"num_hidden_layers": 6,
|
| 28 |
+
"pad_token_id": 0,
|
| 29 |
+
"position_embedding_type": "absolute",
|
| 30 |
+
"problem_type": "single_label_classification",
|
| 31 |
+
"transformers_version": "5.0.0.dev0",
|
| 32 |
+
"type_vocab_size": 2,
|
| 33 |
+
"use_cache": false,
|
| 34 |
+
"vocab_size": 30526
|
| 35 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_accuracy": 0.919369715603382,
|
| 4 |
+
"eval_loss": 0.19649724662303925,
|
| 5 |
+
"eval_runtime": 5.3689,
|
| 6 |
+
"eval_samples": 13010,
|
| 7 |
+
"eval_samples_per_second": 2423.215,
|
| 8 |
+
"eval_steps_per_second": 303.042,
|
| 9 |
+
"num_input_tokens_seen": 33305600
|
| 10 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3997c8da63fc6340b43e6f8d107f48183e754e83933dad0b2aaaf0dc270edd3d
|
| 3 |
+
size 90874096
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<|system|>",
|
| 4 |
+
"<|user|>",
|
| 5 |
+
"<|assistant|>",
|
| 6 |
+
"<|...|>"
|
| 7 |
+
],
|
| 8 |
+
"cls_token": {
|
| 9 |
+
"content": "[CLS]",
|
| 10 |
+
"lstrip": false,
|
| 11 |
+
"normalized": false,
|
| 12 |
+
"rstrip": false,
|
| 13 |
+
"single_word": false
|
| 14 |
+
},
|
| 15 |
+
"mask_token": {
|
| 16 |
+
"content": "[MASK]",
|
| 17 |
+
"lstrip": false,
|
| 18 |
+
"normalized": false,
|
| 19 |
+
"rstrip": false,
|
| 20 |
+
"single_word": false
|
| 21 |
+
},
|
| 22 |
+
"pad_token": {
|
| 23 |
+
"content": "[PAD]",
|
| 24 |
+
"lstrip": false,
|
| 25 |
+
"normalized": false,
|
| 26 |
+
"rstrip": false,
|
| 27 |
+
"single_word": false
|
| 28 |
+
},
|
| 29 |
+
"sep_token": {
|
| 30 |
+
"content": "[SEP]",
|
| 31 |
+
"lstrip": false,
|
| 32 |
+
"normalized": false,
|
| 33 |
+
"rstrip": false,
|
| 34 |
+
"single_word": false
|
| 35 |
+
},
|
| 36 |
+
"unk_token": {
|
| 37 |
+
"content": "[UNK]",
|
| 38 |
+
"lstrip": false,
|
| 39 |
+
"normalized": false,
|
| 40 |
+
"rstrip": false,
|
| 41 |
+
"single_word": false
|
| 42 |
+
}
|
| 43 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,103 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"30522": {
|
| 44 |
+
"content": "<|system|>",
|
| 45 |
+
"lstrip": false,
|
| 46 |
+
"normalized": false,
|
| 47 |
+
"rstrip": false,
|
| 48 |
+
"single_word": false,
|
| 49 |
+
"special": true
|
| 50 |
+
},
|
| 51 |
+
"30523": {
|
| 52 |
+
"content": "<|user|>",
|
| 53 |
+
"lstrip": false,
|
| 54 |
+
"normalized": false,
|
| 55 |
+
"rstrip": false,
|
| 56 |
+
"single_word": false,
|
| 57 |
+
"special": true
|
| 58 |
+
},
|
| 59 |
+
"30524": {
|
| 60 |
+
"content": "<|assistant|>",
|
| 61 |
+
"lstrip": false,
|
| 62 |
+
"normalized": false,
|
| 63 |
+
"rstrip": false,
|
| 64 |
+
"single_word": false,
|
| 65 |
+
"special": true
|
| 66 |
+
},
|
| 67 |
+
"30525": {
|
| 68 |
+
"content": "<|...|>",
|
| 69 |
+
"lstrip": false,
|
| 70 |
+
"normalized": false,
|
| 71 |
+
"rstrip": false,
|
| 72 |
+
"single_word": false,
|
| 73 |
+
"special": true
|
| 74 |
+
}
|
| 75 |
+
},
|
| 76 |
+
"additional_special_tokens": [
|
| 77 |
+
"<|system|>",
|
| 78 |
+
"<|user|>",
|
| 79 |
+
"<|assistant|>",
|
| 80 |
+
"<|...|>"
|
| 81 |
+
],
|
| 82 |
+
"clean_up_tokenization_spaces": true,
|
| 83 |
+
"cls_token": "[CLS]",
|
| 84 |
+
"do_basic_tokenize": true,
|
| 85 |
+
"do_lower_case": true,
|
| 86 |
+
"extra_special_tokens": {},
|
| 87 |
+
"mask_token": "[MASK]",
|
| 88 |
+
"max_length": 512,
|
| 89 |
+
"model_max_length": 512,
|
| 90 |
+
"never_split": null,
|
| 91 |
+
"pad_to_multiple_of": null,
|
| 92 |
+
"pad_token": "[PAD]",
|
| 93 |
+
"pad_token_type_id": 0,
|
| 94 |
+
"padding_side": "right",
|
| 95 |
+
"sep_token": "[SEP]",
|
| 96 |
+
"stride": 0,
|
| 97 |
+
"strip_accents": null,
|
| 98 |
+
"tokenize_chinese_chars": true,
|
| 99 |
+
"tokenizer_class": "BertTokenizer",
|
| 100 |
+
"truncation_side": "right",
|
| 101 |
+
"truncation_strategy": "longest_first",
|
| 102 |
+
"unk_token": "[UNK]"
|
| 103 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"num_input_tokens_seen": 33305600,
|
| 4 |
+
"total_flos": 2157435918643200.0,
|
| 5 |
+
"train_loss": 0.1371641572881533,
|
| 6 |
+
"train_runtime": 362.4532,
|
| 7 |
+
"train_samples": 52040,
|
| 8 |
+
"train_samples_per_second": 717.886,
|
| 9 |
+
"train_steps_per_second": 89.736
|
| 10 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,744 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 6505,
|
| 3 |
+
"best_metric": 0.19649724662303925,
|
| 4 |
+
"best_model_checkpoint": "/media/user/Expansion1/snowflake-arctic-embed-xs-refusal/checkpoint-6505",
|
| 5 |
+
"epoch": 5.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 32525,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.07686395080707148,
|
| 14 |
+
"grad_norm": 2.155881881713867,
|
| 15 |
+
"learning_rate": 4.923289777094543e-05,
|
| 16 |
+
"loss": 0.3511,
|
| 17 |
+
"num_input_tokens_seen": 512000,
|
| 18 |
+
"step": 500,
|
| 19 |
+
"train_runtime": 5.3031,
|
| 20 |
+
"train_tokens_per_second": 96546.489
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.15372790161414296,
|
| 24 |
+
"grad_norm": 0.6236560344696045,
|
| 25 |
+
"learning_rate": 4.846425826287471e-05,
|
| 26 |
+
"loss": 0.2593,
|
| 27 |
+
"num_input_tokens_seen": 1024000,
|
| 28 |
+
"step": 1000,
|
| 29 |
+
"train_runtime": 10.3436,
|
| 30 |
+
"train_tokens_per_second": 98998.124
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.23059185242121444,
|
| 34 |
+
"grad_norm": 3.8176653385162354,
|
| 35 |
+
"learning_rate": 4.7695618754804e-05,
|
| 36 |
+
"loss": 0.2546,
|
| 37 |
+
"num_input_tokens_seen": 1536000,
|
| 38 |
+
"step": 1500,
|
| 39 |
+
"train_runtime": 15.3626,
|
| 40 |
+
"train_tokens_per_second": 99983.128
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 0.3074558032282859,
|
| 44 |
+
"grad_norm": 2.538367986679077,
|
| 45 |
+
"learning_rate": 4.692697924673328e-05,
|
| 46 |
+
"loss": 0.2377,
|
| 47 |
+
"num_input_tokens_seen": 2048000,
|
| 48 |
+
"step": 2000,
|
| 49 |
+
"train_runtime": 20.351,
|
| 50 |
+
"train_tokens_per_second": 100634.117
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.3843197540353574,
|
| 54 |
+
"grad_norm": 3.922593832015991,
|
| 55 |
+
"learning_rate": 4.615833973866257e-05,
|
| 56 |
+
"loss": 0.2411,
|
| 57 |
+
"num_input_tokens_seen": 2560000,
|
| 58 |
+
"step": 2500,
|
| 59 |
+
"train_runtime": 25.363,
|
| 60 |
+
"train_tokens_per_second": 100934.428
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 0.4611837048424289,
|
| 64 |
+
"grad_norm": 0.728330135345459,
|
| 65 |
+
"learning_rate": 4.5389700230591855e-05,
|
| 66 |
+
"loss": 0.2278,
|
| 67 |
+
"num_input_tokens_seen": 3072000,
|
| 68 |
+
"step": 3000,
|
| 69 |
+
"train_runtime": 30.387,
|
| 70 |
+
"train_tokens_per_second": 101095.729
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 0.5380476556495004,
|
| 74 |
+
"grad_norm": 0.5299816131591797,
|
| 75 |
+
"learning_rate": 4.462106072252114e-05,
|
| 76 |
+
"loss": 0.2422,
|
| 77 |
+
"num_input_tokens_seen": 3584000,
|
| 78 |
+
"step": 3500,
|
| 79 |
+
"train_runtime": 35.3784,
|
| 80 |
+
"train_tokens_per_second": 101304.819
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.6149116064565718,
|
| 84 |
+
"grad_norm": 9.030647277832031,
|
| 85 |
+
"learning_rate": 4.3852421214450424e-05,
|
| 86 |
+
"loss": 0.2269,
|
| 87 |
+
"num_input_tokens_seen": 4096000,
|
| 88 |
+
"step": 4000,
|
| 89 |
+
"train_runtime": 40.3905,
|
| 90 |
+
"train_tokens_per_second": 101409.866
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.6917755572636434,
|
| 94 |
+
"grad_norm": 1.8069450855255127,
|
| 95 |
+
"learning_rate": 4.308378170637971e-05,
|
| 96 |
+
"loss": 0.2378,
|
| 97 |
+
"num_input_tokens_seen": 4608000,
|
| 98 |
+
"step": 4500,
|
| 99 |
+
"train_runtime": 45.4094,
|
| 100 |
+
"train_tokens_per_second": 101476.838
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.7686395080707148,
|
| 104 |
+
"grad_norm": 2.4024786949157715,
|
| 105 |
+
"learning_rate": 4.231514219830899e-05,
|
| 106 |
+
"loss": 0.2129,
|
| 107 |
+
"num_input_tokens_seen": 5120000,
|
| 108 |
+
"step": 5000,
|
| 109 |
+
"train_runtime": 50.4114,
|
| 110 |
+
"train_tokens_per_second": 101564.317
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.8455034588777863,
|
| 114 |
+
"grad_norm": 0.8753976821899414,
|
| 115 |
+
"learning_rate": 4.1546502690238284e-05,
|
| 116 |
+
"loss": 0.2354,
|
| 117 |
+
"num_input_tokens_seen": 5632000,
|
| 118 |
+
"step": 5500,
|
| 119 |
+
"train_runtime": 55.4722,
|
| 120 |
+
"train_tokens_per_second": 101528.336
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 0.9223674096848578,
|
| 124 |
+
"grad_norm": 4.247986316680908,
|
| 125 |
+
"learning_rate": 4.077786318216756e-05,
|
| 126 |
+
"loss": 0.237,
|
| 127 |
+
"num_input_tokens_seen": 6144000,
|
| 128 |
+
"step": 6000,
|
| 129 |
+
"train_runtime": 60.5329,
|
| 130 |
+
"train_tokens_per_second": 101498.514
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.9992313604919293,
|
| 134 |
+
"grad_norm": 7.1489973068237305,
|
| 135 |
+
"learning_rate": 4.000922367409685e-05,
|
| 136 |
+
"loss": 0.2286,
|
| 137 |
+
"num_input_tokens_seen": 6656000,
|
| 138 |
+
"step": 6500,
|
| 139 |
+
"train_runtime": 65.5361,
|
| 140 |
+
"train_tokens_per_second": 101562.357
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 1.0,
|
| 144 |
+
"eval_accuracy": 0.919369715603382,
|
| 145 |
+
"eval_loss": 0.19649724662303925,
|
| 146 |
+
"eval_runtime": 5.1167,
|
| 147 |
+
"eval_samples_per_second": 2542.665,
|
| 148 |
+
"eval_steps_per_second": 317.98,
|
| 149 |
+
"num_input_tokens_seen": 6661120,
|
| 150 |
+
"step": 6505
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 1.0760953112990008,
|
| 154 |
+
"grad_norm": 10.231003761291504,
|
| 155 |
+
"learning_rate": 3.9240584166026136e-05,
|
| 156 |
+
"loss": 0.1751,
|
| 157 |
+
"num_input_tokens_seen": 7168000,
|
| 158 |
+
"step": 7000,
|
| 159 |
+
"train_runtime": 75.9746,
|
| 160 |
+
"train_tokens_per_second": 94347.276
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 1.1529592621060722,
|
| 164 |
+
"grad_norm": 0.037300433963537216,
|
| 165 |
+
"learning_rate": 3.847194465795542e-05,
|
| 166 |
+
"loss": 0.1739,
|
| 167 |
+
"num_input_tokens_seen": 7680000,
|
| 168 |
+
"step": 7500,
|
| 169 |
+
"train_runtime": 81.0147,
|
| 170 |
+
"train_tokens_per_second": 94797.624
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 1.2298232129131437,
|
| 174 |
+
"grad_norm": 10.88604736328125,
|
| 175 |
+
"learning_rate": 3.7703305149884705e-05,
|
| 176 |
+
"loss": 0.1923,
|
| 177 |
+
"num_input_tokens_seen": 8192000,
|
| 178 |
+
"step": 8000,
|
| 179 |
+
"train_runtime": 86.0728,
|
| 180 |
+
"train_tokens_per_second": 95175.201
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 1.3066871637202153,
|
| 184 |
+
"grad_norm": 14.25737190246582,
|
| 185 |
+
"learning_rate": 3.693466564181399e-05,
|
| 186 |
+
"loss": 0.1809,
|
| 187 |
+
"num_input_tokens_seen": 8704000,
|
| 188 |
+
"step": 8500,
|
| 189 |
+
"train_runtime": 91.0822,
|
| 190 |
+
"train_tokens_per_second": 95562.051
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 1.3835511145272867,
|
| 194 |
+
"grad_norm": 0.3044818639755249,
|
| 195 |
+
"learning_rate": 3.6166026133743274e-05,
|
| 196 |
+
"loss": 0.1922,
|
| 197 |
+
"num_input_tokens_seen": 9216000,
|
| 198 |
+
"step": 9000,
|
| 199 |
+
"train_runtime": 96.1257,
|
| 200 |
+
"train_tokens_per_second": 95874.452
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 1.4604150653343582,
|
| 204 |
+
"grad_norm": 15.291511535644531,
|
| 205 |
+
"learning_rate": 3.5397386625672565e-05,
|
| 206 |
+
"loss": 0.171,
|
| 207 |
+
"num_input_tokens_seen": 9728000,
|
| 208 |
+
"step": 9500,
|
| 209 |
+
"train_runtime": 101.1575,
|
| 210 |
+
"train_tokens_per_second": 96166.85
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 1.5372790161414298,
|
| 214 |
+
"grad_norm": 12.030097007751465,
|
| 215 |
+
"learning_rate": 3.462874711760184e-05,
|
| 216 |
+
"loss": 0.1779,
|
| 217 |
+
"num_input_tokens_seen": 10240000,
|
| 218 |
+
"step": 10000,
|
| 219 |
+
"train_runtime": 106.2268,
|
| 220 |
+
"train_tokens_per_second": 96397.497
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 1.614142966948501,
|
| 224 |
+
"grad_norm": 0.17212723195552826,
|
| 225 |
+
"learning_rate": 3.386010760953113e-05,
|
| 226 |
+
"loss": 0.176,
|
| 227 |
+
"num_input_tokens_seen": 10752000,
|
| 228 |
+
"step": 10500,
|
| 229 |
+
"train_runtime": 111.2959,
|
| 230 |
+
"train_tokens_per_second": 96607.35
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 1.6910069177555727,
|
| 234 |
+
"grad_norm": 35.843482971191406,
|
| 235 |
+
"learning_rate": 3.309146810146042e-05,
|
| 236 |
+
"loss": 0.1879,
|
| 237 |
+
"num_input_tokens_seen": 11264000,
|
| 238 |
+
"step": 11000,
|
| 239 |
+
"train_runtime": 116.3574,
|
| 240 |
+
"train_tokens_per_second": 96805.219
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 1.767870868562644,
|
| 244 |
+
"grad_norm": 0.055776312947273254,
|
| 245 |
+
"learning_rate": 3.23228285933897e-05,
|
| 246 |
+
"loss": 0.1749,
|
| 247 |
+
"num_input_tokens_seen": 11776000,
|
| 248 |
+
"step": 11500,
|
| 249 |
+
"train_runtime": 121.386,
|
| 250 |
+
"train_tokens_per_second": 97012.845
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 1.8447348193697155,
|
| 254 |
+
"grad_norm": 0.48420748114585876,
|
| 255 |
+
"learning_rate": 3.1554189085318986e-05,
|
| 256 |
+
"loss": 0.1716,
|
| 257 |
+
"num_input_tokens_seen": 12288000,
|
| 258 |
+
"step": 12000,
|
| 259 |
+
"train_runtime": 126.4198,
|
| 260 |
+
"train_tokens_per_second": 97200.001
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 1.9215987701767872,
|
| 264 |
+
"grad_norm": 0.3808608949184418,
|
| 265 |
+
"learning_rate": 3.078554957724827e-05,
|
| 266 |
+
"loss": 0.1819,
|
| 267 |
+
"num_input_tokens_seen": 12800000,
|
| 268 |
+
"step": 12500,
|
| 269 |
+
"train_runtime": 131.456,
|
| 270 |
+
"train_tokens_per_second": 97370.996
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 1.9984627209838586,
|
| 274 |
+
"grad_norm": 0.15483863651752472,
|
| 275 |
+
"learning_rate": 3.0016910069177555e-05,
|
| 276 |
+
"loss": 0.1718,
|
| 277 |
+
"num_input_tokens_seen": 13312000,
|
| 278 |
+
"step": 13000,
|
| 279 |
+
"train_runtime": 136.5089,
|
| 280 |
+
"train_tokens_per_second": 97517.416
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 2.0,
|
| 284 |
+
"eval_accuracy": 0.9259031514219831,
|
| 285 |
+
"eval_loss": 0.27698734402656555,
|
| 286 |
+
"eval_runtime": 5.1343,
|
| 287 |
+
"eval_samples_per_second": 2533.949,
|
| 288 |
+
"eval_steps_per_second": 316.89,
|
| 289 |
+
"num_input_tokens_seen": 13322240,
|
| 290 |
+
"step": 13010
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 2.07532667179093,
|
| 294 |
+
"grad_norm": 0.06390306353569031,
|
| 295 |
+
"learning_rate": 2.9248270561106846e-05,
|
| 296 |
+
"loss": 0.1384,
|
| 297 |
+
"num_input_tokens_seen": 13824000,
|
| 298 |
+
"step": 13500,
|
| 299 |
+
"train_runtime": 146.9662,
|
| 300 |
+
"train_tokens_per_second": 94062.42
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 2.1521906225980016,
|
| 304 |
+
"grad_norm": 0.09185440093278885,
|
| 305 |
+
"learning_rate": 2.8479631053036127e-05,
|
| 306 |
+
"loss": 0.123,
|
| 307 |
+
"num_input_tokens_seen": 14336000,
|
| 308 |
+
"step": 14000,
|
| 309 |
+
"train_runtime": 152.0064,
|
| 310 |
+
"train_tokens_per_second": 94311.832
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 2.229054573405073,
|
| 314 |
+
"grad_norm": 0.05354034900665283,
|
| 315 |
+
"learning_rate": 2.7710991544965414e-05,
|
| 316 |
+
"loss": 0.1265,
|
| 317 |
+
"num_input_tokens_seen": 14848000,
|
| 318 |
+
"step": 14500,
|
| 319 |
+
"train_runtime": 157.0491,
|
| 320 |
+
"train_tokens_per_second": 94543.682
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 2.3059185242121445,
|
| 324 |
+
"grad_norm": 6.2592926025390625,
|
| 325 |
+
"learning_rate": 2.6942352036894695e-05,
|
| 326 |
+
"loss": 0.1214,
|
| 327 |
+
"num_input_tokens_seen": 15360000,
|
| 328 |
+
"step": 15000,
|
| 329 |
+
"train_runtime": 162.0991,
|
| 330 |
+
"train_tokens_per_second": 94756.854
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 2.382782475019216,
|
| 334 |
+
"grad_norm": 0.3284030854701996,
|
| 335 |
+
"learning_rate": 2.6173712528823986e-05,
|
| 336 |
+
"loss": 0.1298,
|
| 337 |
+
"num_input_tokens_seen": 15872000,
|
| 338 |
+
"step": 15500,
|
| 339 |
+
"train_runtime": 167.141,
|
| 340 |
+
"train_tokens_per_second": 94961.759
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 2.4596464258262873,
|
| 344 |
+
"grad_norm": 0.2101190984249115,
|
| 345 |
+
"learning_rate": 2.5405073020753267e-05,
|
| 346 |
+
"loss": 0.1279,
|
| 347 |
+
"num_input_tokens_seen": 16384000,
|
| 348 |
+
"step": 16000,
|
| 349 |
+
"train_runtime": 172.1787,
|
| 350 |
+
"train_tokens_per_second": 95156.958
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 2.536510376633359,
|
| 354 |
+
"grad_norm": 7.672014236450195,
|
| 355 |
+
"learning_rate": 2.463643351268255e-05,
|
| 356 |
+
"loss": 0.1435,
|
| 357 |
+
"num_input_tokens_seen": 16896000,
|
| 358 |
+
"step": 16500,
|
| 359 |
+
"train_runtime": 177.2341,
|
| 360 |
+
"train_tokens_per_second": 95331.566
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 2.6133743274404306,
|
| 364 |
+
"grad_norm": 37.9052734375,
|
| 365 |
+
"learning_rate": 2.3867794004611836e-05,
|
| 366 |
+
"loss": 0.123,
|
| 367 |
+
"num_input_tokens_seen": 17408000,
|
| 368 |
+
"step": 17000,
|
| 369 |
+
"train_runtime": 182.2827,
|
| 370 |
+
"train_tokens_per_second": 95500.03
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 2.690238278247502,
|
| 374 |
+
"grad_norm": 0.08578933030366898,
|
| 375 |
+
"learning_rate": 2.3099154496541124e-05,
|
| 376 |
+
"loss": 0.1289,
|
| 377 |
+
"num_input_tokens_seen": 17920000,
|
| 378 |
+
"step": 17500,
|
| 379 |
+
"train_runtime": 187.3355,
|
| 380 |
+
"train_tokens_per_second": 95657.272
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 2.7671022290545735,
|
| 384 |
+
"grad_norm": 0.08860859274864197,
|
| 385 |
+
"learning_rate": 2.2330514988470408e-05,
|
| 386 |
+
"loss": 0.1296,
|
| 387 |
+
"num_input_tokens_seen": 18432000,
|
| 388 |
+
"step": 18000,
|
| 389 |
+
"train_runtime": 192.3781,
|
| 390 |
+
"train_tokens_per_second": 95811.329
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 2.8439661798616447,
|
| 394 |
+
"grad_norm": 0.41104796528816223,
|
| 395 |
+
"learning_rate": 2.1561875480399692e-05,
|
| 396 |
+
"loss": 0.1124,
|
| 397 |
+
"num_input_tokens_seen": 18944000,
|
| 398 |
+
"step": 18500,
|
| 399 |
+
"train_runtime": 197.7595,
|
| 400 |
+
"train_tokens_per_second": 95793.134
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 2.9208301306687163,
|
| 404 |
+
"grad_norm": 37.97283172607422,
|
| 405 |
+
"learning_rate": 2.079323597232898e-05,
|
| 406 |
+
"loss": 0.1382,
|
| 407 |
+
"num_input_tokens_seen": 19456000,
|
| 408 |
+
"step": 19000,
|
| 409 |
+
"train_runtime": 203.2967,
|
| 410 |
+
"train_tokens_per_second": 95702.474
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 2.997694081475788,
|
| 414 |
+
"grad_norm": 0.0325402170419693,
|
| 415 |
+
"learning_rate": 2.0024596464258264e-05,
|
| 416 |
+
"loss": 0.1388,
|
| 417 |
+
"num_input_tokens_seen": 19968000,
|
| 418 |
+
"step": 19500,
|
| 419 |
+
"train_runtime": 208.8029,
|
| 420 |
+
"train_tokens_per_second": 95630.843
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 3.0,
|
| 424 |
+
"eval_accuracy": 0.9287471176018447,
|
| 425 |
+
"eval_loss": 0.31825903058052063,
|
| 426 |
+
"eval_runtime": 5.4038,
|
| 427 |
+
"eval_samples_per_second": 2407.562,
|
| 428 |
+
"eval_steps_per_second": 301.084,
|
| 429 |
+
"num_input_tokens_seen": 19983360,
|
| 430 |
+
"step": 19515
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 3.074558032282859,
|
| 434 |
+
"grad_norm": 0.05180477350950241,
|
| 435 |
+
"learning_rate": 1.925595695618755e-05,
|
| 436 |
+
"loss": 0.0823,
|
| 437 |
+
"num_input_tokens_seen": 20480000,
|
| 438 |
+
"step": 20000,
|
| 439 |
+
"train_runtime": 219.9739,
|
| 440 |
+
"train_tokens_per_second": 93101.956
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 3.151421983089931,
|
| 444 |
+
"grad_norm": 0.005111335311084986,
|
| 445 |
+
"learning_rate": 1.8487317448116833e-05,
|
| 446 |
+
"loss": 0.0716,
|
| 447 |
+
"num_input_tokens_seen": 20992000,
|
| 448 |
+
"step": 20500,
|
| 449 |
+
"train_runtime": 225.3567,
|
| 450 |
+
"train_tokens_per_second": 93150.099
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 3.2282859338970025,
|
| 454 |
+
"grad_norm": 0.012623129412531853,
|
| 455 |
+
"learning_rate": 1.771867794004612e-05,
|
| 456 |
+
"loss": 0.0915,
|
| 457 |
+
"num_input_tokens_seen": 21504000,
|
| 458 |
+
"step": 21000,
|
| 459 |
+
"train_runtime": 230.6352,
|
| 460 |
+
"train_tokens_per_second": 93238.15
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 3.3051498847040737,
|
| 464 |
+
"grad_norm": 10.89956283569336,
|
| 465 |
+
"learning_rate": 1.6950038431975405e-05,
|
| 466 |
+
"loss": 0.0783,
|
| 467 |
+
"num_input_tokens_seen": 22016000,
|
| 468 |
+
"step": 21500,
|
| 469 |
+
"train_runtime": 236.059,
|
| 470 |
+
"train_tokens_per_second": 93264.831
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 3.3820138355111453,
|
| 474 |
+
"grad_norm": 0.010125258006155491,
|
| 475 |
+
"learning_rate": 1.618139892390469e-05,
|
| 476 |
+
"loss": 0.0882,
|
| 477 |
+
"num_input_tokens_seen": 22528000,
|
| 478 |
+
"step": 22000,
|
| 479 |
+
"train_runtime": 241.5242,
|
| 480 |
+
"train_tokens_per_second": 93274.31
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 3.458877786318217,
|
| 484 |
+
"grad_norm": 0.04097803309559822,
|
| 485 |
+
"learning_rate": 1.5412759415833973e-05,
|
| 486 |
+
"loss": 0.095,
|
| 487 |
+
"num_input_tokens_seen": 23040000,
|
| 488 |
+
"step": 22500,
|
| 489 |
+
"train_runtime": 247.0361,
|
| 490 |
+
"train_tokens_per_second": 93265.737
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 3.535741737125288,
|
| 494 |
+
"grad_norm": 0.3172767758369446,
|
| 495 |
+
"learning_rate": 1.464411990776326e-05,
|
| 496 |
+
"loss": 0.0847,
|
| 497 |
+
"num_input_tokens_seen": 23552000,
|
| 498 |
+
"step": 23000,
|
| 499 |
+
"train_runtime": 252.366,
|
| 500 |
+
"train_tokens_per_second": 93324.783
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 3.61260568793236,
|
| 504 |
+
"grad_norm": 0.0049354820512235165,
|
| 505 |
+
"learning_rate": 1.3875480399692545e-05,
|
| 506 |
+
"loss": 0.0758,
|
| 507 |
+
"num_input_tokens_seen": 24064000,
|
| 508 |
+
"step": 23500,
|
| 509 |
+
"train_runtime": 257.7685,
|
| 510 |
+
"train_tokens_per_second": 93355.1
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 3.689469638739431,
|
| 514 |
+
"grad_norm": 0.09353843331336975,
|
| 515 |
+
"learning_rate": 1.310684089162183e-05,
|
| 516 |
+
"loss": 0.0812,
|
| 517 |
+
"num_input_tokens_seen": 24576000,
|
| 518 |
+
"step": 24000,
|
| 519 |
+
"train_runtime": 263.0331,
|
| 520 |
+
"train_tokens_per_second": 93433.102
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 3.7663335895465027,
|
| 524 |
+
"grad_norm": 13.593195915222168,
|
| 525 |
+
"learning_rate": 1.2338201383551116e-05,
|
| 526 |
+
"loss": 0.0882,
|
| 527 |
+
"num_input_tokens_seen": 25088000,
|
| 528 |
+
"step": 24500,
|
| 529 |
+
"train_runtime": 268.3188,
|
| 530 |
+
"train_tokens_per_second": 93500.701
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 3.8431975403535743,
|
| 534 |
+
"grad_norm": 0.12645399570465088,
|
| 535 |
+
"learning_rate": 1.15695618754804e-05,
|
| 536 |
+
"loss": 0.0862,
|
| 537 |
+
"num_input_tokens_seen": 25600000,
|
| 538 |
+
"step": 25000,
|
| 539 |
+
"train_runtime": 273.6907,
|
| 540 |
+
"train_tokens_per_second": 93536.259
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 3.9200614911606455,
|
| 544 |
+
"grad_norm": 0.015061162412166595,
|
| 545 |
+
"learning_rate": 1.0800922367409686e-05,
|
| 546 |
+
"loss": 0.0825,
|
| 547 |
+
"num_input_tokens_seen": 26112000,
|
| 548 |
+
"step": 25500,
|
| 549 |
+
"train_runtime": 279.0408,
|
| 550 |
+
"train_tokens_per_second": 93577.699
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 3.996925441967717,
|
| 554 |
+
"grad_norm": 0.020776506513357162,
|
| 555 |
+
"learning_rate": 1.003228285933897e-05,
|
| 556 |
+
"loss": 0.0772,
|
| 557 |
+
"num_input_tokens_seen": 26624000,
|
| 558 |
+
"step": 26000,
|
| 559 |
+
"train_runtime": 284.4194,
|
| 560 |
+
"train_tokens_per_second": 93608.25
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 4.0,
|
| 564 |
+
"eval_accuracy": 0.9270561106840891,
|
| 565 |
+
"eval_loss": 0.43025368452072144,
|
| 566 |
+
"eval_runtime": 5.3244,
|
| 567 |
+
"eval_samples_per_second": 2443.487,
|
| 568 |
+
"eval_steps_per_second": 305.577,
|
| 569 |
+
"num_input_tokens_seen": 26644480,
|
| 570 |
+
"step": 26020
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 4.073789392774788,
|
| 574 |
+
"grad_norm": 0.09127756953239441,
|
| 575 |
+
"learning_rate": 9.263643351268256e-06,
|
| 576 |
+
"loss": 0.0615,
|
| 577 |
+
"num_input_tokens_seen": 27136000,
|
| 578 |
+
"step": 26500,
|
| 579 |
+
"train_runtime": 295.1257,
|
| 580 |
+
"train_tokens_per_second": 91947.265
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"epoch": 4.15065334358186,
|
| 584 |
+
"grad_norm": 0.11556842923164368,
|
| 585 |
+
"learning_rate": 8.49500384319754e-06,
|
| 586 |
+
"loss": 0.0444,
|
| 587 |
+
"num_input_tokens_seen": 27648000,
|
| 588 |
+
"step": 27000,
|
| 589 |
+
"train_runtime": 300.149,
|
| 590 |
+
"train_tokens_per_second": 92114.259
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 4.227517294388932,
|
| 594 |
+
"grad_norm": 0.09005430340766907,
|
| 595 |
+
"learning_rate": 7.726364335126826e-06,
|
| 596 |
+
"loss": 0.0434,
|
| 597 |
+
"num_input_tokens_seen": 28160000,
|
| 598 |
+
"step": 27500,
|
| 599 |
+
"train_runtime": 305.2327,
|
| 600 |
+
"train_tokens_per_second": 92257.495
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 4.304381245196003,
|
| 604 |
+
"grad_norm": 0.004569609649479389,
|
| 605 |
+
"learning_rate": 6.9577248270561115e-06,
|
| 606 |
+
"loss": 0.0352,
|
| 607 |
+
"num_input_tokens_seen": 28672000,
|
| 608 |
+
"step": 28000,
|
| 609 |
+
"train_runtime": 310.2962,
|
| 610 |
+
"train_tokens_per_second": 92402.026
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 4.381245196003075,
|
| 614 |
+
"grad_norm": 1.524936318397522,
|
| 615 |
+
"learning_rate": 6.189085318985397e-06,
|
| 616 |
+
"loss": 0.06,
|
| 617 |
+
"num_input_tokens_seen": 29184000,
|
| 618 |
+
"step": 28500,
|
| 619 |
+
"train_runtime": 315.365,
|
| 620 |
+
"train_tokens_per_second": 92540.388
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 4.458109146810146,
|
| 624 |
+
"grad_norm": 0.014427268877625465,
|
| 625 |
+
"learning_rate": 5.420445810914681e-06,
|
| 626 |
+
"loss": 0.0492,
|
| 627 |
+
"num_input_tokens_seen": 29696000,
|
| 628 |
+
"step": 29000,
|
| 629 |
+
"train_runtime": 320.4274,
|
| 630 |
+
"train_tokens_per_second": 92676.208
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 4.534973097617217,
|
| 634 |
+
"grad_norm": 0.07355033606290817,
|
| 635 |
+
"learning_rate": 4.651806302843966e-06,
|
| 636 |
+
"loss": 0.0572,
|
| 637 |
+
"num_input_tokens_seen": 30208000,
|
| 638 |
+
"step": 29500,
|
| 639 |
+
"train_runtime": 325.4966,
|
| 640 |
+
"train_tokens_per_second": 92805.883
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 4.611837048424289,
|
| 644 |
+
"grad_norm": 7.176478385925293,
|
| 645 |
+
"learning_rate": 3.883166794773251e-06,
|
| 646 |
+
"loss": 0.0442,
|
| 647 |
+
"num_input_tokens_seen": 30720000,
|
| 648 |
+
"step": 30000,
|
| 649 |
+
"train_runtime": 330.568,
|
| 650 |
+
"train_tokens_per_second": 92930.968
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 4.688700999231361,
|
| 654 |
+
"grad_norm": 0.030576860532164574,
|
| 655 |
+
"learning_rate": 3.114527286702537e-06,
|
| 656 |
+
"loss": 0.0465,
|
| 657 |
+
"num_input_tokens_seen": 31232000,
|
| 658 |
+
"step": 30500,
|
| 659 |
+
"train_runtime": 335.6493,
|
| 660 |
+
"train_tokens_per_second": 93049.488
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 4.765564950038432,
|
| 664 |
+
"grad_norm": 0.005597515497356653,
|
| 665 |
+
"learning_rate": 2.345887778631822e-06,
|
| 666 |
+
"loss": 0.0487,
|
| 667 |
+
"num_input_tokens_seen": 31744000,
|
| 668 |
+
"step": 31000,
|
| 669 |
+
"train_runtime": 340.8069,
|
| 670 |
+
"train_tokens_per_second": 93143.641
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 4.842428900845503,
|
| 674 |
+
"grad_norm": 0.021089155226945877,
|
| 675 |
+
"learning_rate": 1.5772482705611067e-06,
|
| 676 |
+
"loss": 0.0557,
|
| 677 |
+
"num_input_tokens_seen": 32256000,
|
| 678 |
+
"step": 31500,
|
| 679 |
+
"train_runtime": 346.0097,
|
| 680 |
+
"train_tokens_per_second": 93222.824
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 4.919292851652575,
|
| 684 |
+
"grad_norm": 0.023516027256846428,
|
| 685 |
+
"learning_rate": 8.086087624903922e-07,
|
| 686 |
+
"loss": 0.0399,
|
| 687 |
+
"num_input_tokens_seen": 32768000,
|
| 688 |
+
"step": 32000,
|
| 689 |
+
"train_runtime": 351.2595,
|
| 690 |
+
"train_tokens_per_second": 93287.165
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 4.996156802459646,
|
| 694 |
+
"grad_norm": 0.007581554353237152,
|
| 695 |
+
"learning_rate": 3.996925441967718e-08,
|
| 696 |
+
"loss": 0.0532,
|
| 697 |
+
"num_input_tokens_seen": 33280000,
|
| 698 |
+
"step": 32500,
|
| 699 |
+
"train_runtime": 356.5514,
|
| 700 |
+
"train_tokens_per_second": 93338.566
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"epoch": 5.0,
|
| 704 |
+
"eval_accuracy": 0.9264411990776326,
|
| 705 |
+
"eval_loss": 0.4613732397556305,
|
| 706 |
+
"eval_runtime": 5.2997,
|
| 707 |
+
"eval_samples_per_second": 2454.857,
|
| 708 |
+
"eval_steps_per_second": 306.999,
|
| 709 |
+
"num_input_tokens_seen": 33305600,
|
| 710 |
+
"step": 32525
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 5.0,
|
| 714 |
+
"num_input_tokens_seen": 33305600,
|
| 715 |
+
"step": 32525,
|
| 716 |
+
"total_flos": 2157435918643200.0,
|
| 717 |
+
"train_loss": 0.1371641572881533,
|
| 718 |
+
"train_runtime": 362.4532,
|
| 719 |
+
"train_samples_per_second": 717.886,
|
| 720 |
+
"train_steps_per_second": 89.736
|
| 721 |
+
}
|
| 722 |
+
],
|
| 723 |
+
"logging_steps": 500,
|
| 724 |
+
"max_steps": 32525,
|
| 725 |
+
"num_input_tokens_seen": 33305600,
|
| 726 |
+
"num_train_epochs": 5,
|
| 727 |
+
"save_steps": 500,
|
| 728 |
+
"stateful_callbacks": {
|
| 729 |
+
"TrainerControl": {
|
| 730 |
+
"args": {
|
| 731 |
+
"should_epoch_stop": false,
|
| 732 |
+
"should_evaluate": false,
|
| 733 |
+
"should_log": false,
|
| 734 |
+
"should_save": true,
|
| 735 |
+
"should_training_stop": true
|
| 736 |
+
},
|
| 737 |
+
"attributes": {}
|
| 738 |
+
}
|
| 739 |
+
},
|
| 740 |
+
"total_flos": 2157435918643200.0,
|
| 741 |
+
"train_batch_size": 8,
|
| 742 |
+
"trial_name": null,
|
| 743 |
+
"trial_params": null
|
| 744 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:88bc931c3ffc6b4b25fa2a994558f90c9fe24e5f81f4d486bf8c879b0f031021
|
| 3 |
+
size 5201
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|