Training in progress, step 6000, checkpoint
Browse files- checkpoint-6000/config.json +30 -0
- checkpoint-6000/generation_config.json +6 -0
- checkpoint-6000/model.safetensors +3 -0
- checkpoint-6000/optimizer.pt +3 -0
- checkpoint-6000/rng_state.pth +3 -0
- checkpoint-6000/scheduler.pt +3 -0
- checkpoint-6000/special_tokens_map.json +24 -0
- checkpoint-6000/tokenizer.json +96 -0
- checkpoint-6000/tokenizer_config.json +38 -0
- checkpoint-6000/trainer_state.json +453 -0
- checkpoint-6000/training_args.bin +3 -0
checkpoint-6000/config.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "EleutherAI/pythia-70m",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"GPTNeoXForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": true,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 0,
|
| 9 |
+
"classifier_dropout": 0.1,
|
| 10 |
+
"eos_token_id": 0,
|
| 11 |
+
"hidden_act": "gelu",
|
| 12 |
+
"hidden_dropout": 0.0,
|
| 13 |
+
"hidden_size": 512,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 2048,
|
| 16 |
+
"layer_norm_eps": 1e-05,
|
| 17 |
+
"max_position_embeddings": 2048,
|
| 18 |
+
"model_type": "gpt_neox",
|
| 19 |
+
"num_attention_heads": 8,
|
| 20 |
+
"num_hidden_layers": 6,
|
| 21 |
+
"rope_scaling": null,
|
| 22 |
+
"rotary_emb_base": 10000,
|
| 23 |
+
"rotary_pct": 0.25,
|
| 24 |
+
"tie_word_embeddings": false,
|
| 25 |
+
"torch_dtype": "float32",
|
| 26 |
+
"transformers_version": "4.42.3",
|
| 27 |
+
"use_cache": true,
|
| 28 |
+
"use_parallel_residual": true,
|
| 29 |
+
"vocab_size": 8
|
| 30 |
+
}
|
checkpoint-6000/generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 0,
|
| 4 |
+
"eos_token_id": 0,
|
| 5 |
+
"transformers_version": "4.42.3"
|
| 6 |
+
}
|
checkpoint-6000/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:20eb895add63a221ec00035773514959871ba1812d056d2f65d3fd81f099a8dd
|
| 3 |
+
size 75702552
|
checkpoint-6000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ccaaad72a66f821893d8313040271220a99fa778696a65a8e0fc4a56db3578ce
|
| 3 |
+
size 151452026
|
checkpoint-6000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:835f869ea325fd6edf27b48b589309fb66641cb92b45f2fc13d1bb6e8814106c
|
| 3 |
+
size 14244
|
checkpoint-6000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b2213926e5f1b1c00971ae263217677342a24510c98bb83a84a1627d10ef84a3
|
| 3 |
+
size 1064
|
checkpoint-6000/special_tokens_map.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<|beginoftext|>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": true,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "<|endoftext|>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": "<|endoftext|>",
|
| 17 |
+
"unk_token": {
|
| 18 |
+
"content": "<|unknown|>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": false,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
}
|
| 24 |
+
}
|
checkpoint-6000/tokenizer.json
ADDED
|
@@ -0,0 +1,96 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"version": "1.0",
|
| 3 |
+
"truncation": null,
|
| 4 |
+
"padding": null,
|
| 5 |
+
"added_tokens": [
|
| 6 |
+
{
|
| 7 |
+
"id": 0,
|
| 8 |
+
"content": "<|beginoftext|>",
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"lstrip": false,
|
| 11 |
+
"rstrip": false,
|
| 12 |
+
"normalized": true,
|
| 13 |
+
"special": true
|
| 14 |
+
},
|
| 15 |
+
{
|
| 16 |
+
"id": 1,
|
| 17 |
+
"content": "<|endoftext|>",
|
| 18 |
+
"single_word": false,
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"normalized": false,
|
| 22 |
+
"special": true
|
| 23 |
+
},
|
| 24 |
+
{
|
| 25 |
+
"id": 2,
|
| 26 |
+
"content": "<|unknown|>",
|
| 27 |
+
"single_word": false,
|
| 28 |
+
"lstrip": false,
|
| 29 |
+
"rstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"special": true
|
| 32 |
+
}
|
| 33 |
+
],
|
| 34 |
+
"normalizer": {
|
| 35 |
+
"type": "NFC"
|
| 36 |
+
},
|
| 37 |
+
"pre_tokenizer": {
|
| 38 |
+
"type": "ByteLevel",
|
| 39 |
+
"add_prefix_space": false,
|
| 40 |
+
"trim_offsets": true,
|
| 41 |
+
"use_regex": true
|
| 42 |
+
},
|
| 43 |
+
"post_processor": {
|
| 44 |
+
"type": "TemplateProcessing",
|
| 45 |
+
"single": [
|
| 46 |
+
{
|
| 47 |
+
"Sequence": {
|
| 48 |
+
"id": "A",
|
| 49 |
+
"type_id": 0
|
| 50 |
+
}
|
| 51 |
+
}
|
| 52 |
+
],
|
| 53 |
+
"pair": [
|
| 54 |
+
{
|
| 55 |
+
"Sequence": {
|
| 56 |
+
"id": "A",
|
| 57 |
+
"type_id": 0
|
| 58 |
+
}
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"Sequence": {
|
| 62 |
+
"id": "B",
|
| 63 |
+
"type_id": 1
|
| 64 |
+
}
|
| 65 |
+
}
|
| 66 |
+
],
|
| 67 |
+
"special_tokens": {}
|
| 68 |
+
},
|
| 69 |
+
"decoder": {
|
| 70 |
+
"type": "ByteLevel",
|
| 71 |
+
"add_prefix_space": false,
|
| 72 |
+
"trim_offsets": true,
|
| 73 |
+
"use_regex": true
|
| 74 |
+
},
|
| 75 |
+
"model": {
|
| 76 |
+
"type": "BPE",
|
| 77 |
+
"dropout": null,
|
| 78 |
+
"unk_token": "<|unknown|>",
|
| 79 |
+
"continuing_subword_prefix": null,
|
| 80 |
+
"end_of_word_suffix": null,
|
| 81 |
+
"fuse_unk": false,
|
| 82 |
+
"byte_fallback": false,
|
| 83 |
+
"ignore_merges": false,
|
| 84 |
+
"vocab": {
|
| 85 |
+
"<|beginoftext|>": 0,
|
| 86 |
+
"<|padding|>": 1,
|
| 87 |
+
"<|unknown|>": 2,
|
| 88 |
+
"a": 3,
|
| 89 |
+
"c": 4,
|
| 90 |
+
"g": 5,
|
| 91 |
+
"t": 6,
|
| 92 |
+
"n": 7
|
| 93 |
+
},
|
| 94 |
+
"merges": []
|
| 95 |
+
}
|
| 96 |
+
}
|
checkpoint-6000/tokenizer_config.json
ADDED
|
@@ -0,0 +1,38 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": false,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"add_prefix_space": false,
|
| 5 |
+
"added_tokens_decoder": {
|
| 6 |
+
"0": {
|
| 7 |
+
"content": "<|beginoftext|>",
|
| 8 |
+
"lstrip": false,
|
| 9 |
+
"normalized": true,
|
| 10 |
+
"rstrip": false,
|
| 11 |
+
"single_word": false,
|
| 12 |
+
"special": true
|
| 13 |
+
},
|
| 14 |
+
"1": {
|
| 15 |
+
"content": "<|endoftext|>",
|
| 16 |
+
"lstrip": false,
|
| 17 |
+
"normalized": false,
|
| 18 |
+
"rstrip": false,
|
| 19 |
+
"single_word": false,
|
| 20 |
+
"special": true
|
| 21 |
+
},
|
| 22 |
+
"2": {
|
| 23 |
+
"content": "<|unknown|>",
|
| 24 |
+
"lstrip": false,
|
| 25 |
+
"normalized": false,
|
| 26 |
+
"rstrip": false,
|
| 27 |
+
"single_word": false,
|
| 28 |
+
"special": true
|
| 29 |
+
}
|
| 30 |
+
},
|
| 31 |
+
"bos_token": "<|beginoftext|>",
|
| 32 |
+
"clean_up_tokenization_spaces": true,
|
| 33 |
+
"eos_token": "<|endoftext|>",
|
| 34 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 35 |
+
"pad_token": "<|endoftext|>",
|
| 36 |
+
"tokenizer_class": "GPTNeoXTokenizer",
|
| 37 |
+
"unk_token": "<|unknown|>"
|
| 38 |
+
}
|
checkpoint-6000/trainer_state.json
ADDED
|
@@ -0,0 +1,453 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.17589118198874296,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 6000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0029315196998123826,
|
| 13 |
+
"grad_norm": 0.7248865962028503,
|
| 14 |
+
"learning_rate": 4.9868042930033434e-05,
|
| 15 |
+
"loss": 1.3593,
|
| 16 |
+
"step": 100
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.005863039399624765,
|
| 20 |
+
"grad_norm": 0.3698488771915436,
|
| 21 |
+
"learning_rate": 4.972142396340391e-05,
|
| 22 |
+
"loss": 1.3285,
|
| 23 |
+
"step": 200
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.008794559099437148,
|
| 27 |
+
"grad_norm": 0.24848727881908417,
|
| 28 |
+
"learning_rate": 4.957480499677439e-05,
|
| 29 |
+
"loss": 1.324,
|
| 30 |
+
"step": 300
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.01172607879924953,
|
| 34 |
+
"grad_norm": 0.2648352086544037,
|
| 35 |
+
"learning_rate": 4.942818603014486e-05,
|
| 36 |
+
"loss": 1.3145,
|
| 37 |
+
"step": 400
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.014657598499061914,
|
| 41 |
+
"grad_norm": 0.3995356559753418,
|
| 42 |
+
"learning_rate": 4.928156706351534e-05,
|
| 43 |
+
"loss": 1.3141,
|
| 44 |
+
"step": 500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.017589118198874296,
|
| 48 |
+
"grad_norm": 0.32299748063087463,
|
| 49 |
+
"learning_rate": 4.913494809688581e-05,
|
| 50 |
+
"loss": 1.3087,
|
| 51 |
+
"step": 600
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.02052063789868668,
|
| 55 |
+
"grad_norm": 0.33237338066101074,
|
| 56 |
+
"learning_rate": 4.898832913025629e-05,
|
| 57 |
+
"loss": 1.3071,
|
| 58 |
+
"step": 700
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.02345215759849906,
|
| 62 |
+
"grad_norm": 0.2716487646102905,
|
| 63 |
+
"learning_rate": 4.8841710163626766e-05,
|
| 64 |
+
"loss": 1.3048,
|
| 65 |
+
"step": 800
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.026383677298311446,
|
| 69 |
+
"grad_norm": 0.8654017448425293,
|
| 70 |
+
"learning_rate": 4.869509119699725e-05,
|
| 71 |
+
"loss": 1.3087,
|
| 72 |
+
"step": 900
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.02931519699812383,
|
| 76 |
+
"grad_norm": 0.27945375442504883,
|
| 77 |
+
"learning_rate": 4.854847223036773e-05,
|
| 78 |
+
"loss": 1.3015,
|
| 79 |
+
"step": 1000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.03224671669793621,
|
| 83 |
+
"grad_norm": 0.27838173508644104,
|
| 84 |
+
"learning_rate": 4.84018532637382e-05,
|
| 85 |
+
"loss": 1.2998,
|
| 86 |
+
"step": 1100
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.03517823639774859,
|
| 90 |
+
"grad_norm": 0.24157749116420746,
|
| 91 |
+
"learning_rate": 4.8255234297108676e-05,
|
| 92 |
+
"loss": 1.3016,
|
| 93 |
+
"step": 1200
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.038109756097560975,
|
| 97 |
+
"grad_norm": 0.45793822407722473,
|
| 98 |
+
"learning_rate": 4.810861533047915e-05,
|
| 99 |
+
"loss": 1.2983,
|
| 100 |
+
"step": 1300
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.04104127579737336,
|
| 104 |
+
"grad_norm": 0.1312064826488495,
|
| 105 |
+
"learning_rate": 4.796199636384963e-05,
|
| 106 |
+
"loss": 1.2978,
|
| 107 |
+
"step": 1400
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.04397279549718574,
|
| 111 |
+
"grad_norm": 0.33062976598739624,
|
| 112 |
+
"learning_rate": 4.7815377397220105e-05,
|
| 113 |
+
"loss": 1.2972,
|
| 114 |
+
"step": 1500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.04690431519699812,
|
| 118 |
+
"grad_norm": 0.29819390177726746,
|
| 119 |
+
"learning_rate": 4.766875843059058e-05,
|
| 120 |
+
"loss": 1.2945,
|
| 121 |
+
"step": 1600
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.0498358348968105,
|
| 125 |
+
"grad_norm": 0.3396131694316864,
|
| 126 |
+
"learning_rate": 4.752213946396106e-05,
|
| 127 |
+
"loss": 1.2921,
|
| 128 |
+
"step": 1700
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.05276735459662289,
|
| 132 |
+
"grad_norm": 0.3231181502342224,
|
| 133 |
+
"learning_rate": 4.737552049733154e-05,
|
| 134 |
+
"loss": 1.2932,
|
| 135 |
+
"step": 1800
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.055698874296435275,
|
| 139 |
+
"grad_norm": 0.4015660285949707,
|
| 140 |
+
"learning_rate": 4.7228901530702015e-05,
|
| 141 |
+
"loss": 1.2899,
|
| 142 |
+
"step": 1900
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.05863039399624766,
|
| 146 |
+
"grad_norm": 0.436213880777359,
|
| 147 |
+
"learning_rate": 4.708228256407249e-05,
|
| 148 |
+
"loss": 1.2906,
|
| 149 |
+
"step": 2000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.06156191369606004,
|
| 153 |
+
"grad_norm": 0.3451833426952362,
|
| 154 |
+
"learning_rate": 4.693566359744297e-05,
|
| 155 |
+
"loss": 1.2884,
|
| 156 |
+
"step": 2100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.06449343339587242,
|
| 160 |
+
"grad_norm": 0.41890543699264526,
|
| 161 |
+
"learning_rate": 4.6789044630813445e-05,
|
| 162 |
+
"loss": 1.2892,
|
| 163 |
+
"step": 2200
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.0674249530956848,
|
| 167 |
+
"grad_norm": 0.3117181062698364,
|
| 168 |
+
"learning_rate": 4.664242566418392e-05,
|
| 169 |
+
"loss": 1.2866,
|
| 170 |
+
"step": 2300
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.07035647279549719,
|
| 174 |
+
"grad_norm": 0.2703840136528015,
|
| 175 |
+
"learning_rate": 4.64958066975544e-05,
|
| 176 |
+
"loss": 1.2869,
|
| 177 |
+
"step": 2400
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.07328799249530957,
|
| 181 |
+
"grad_norm": 0.31400740146636963,
|
| 182 |
+
"learning_rate": 4.6349187730924874e-05,
|
| 183 |
+
"loss": 1.2844,
|
| 184 |
+
"step": 2500
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.07621951219512195,
|
| 188 |
+
"grad_norm": 0.36265355348587036,
|
| 189 |
+
"learning_rate": 4.620256876429535e-05,
|
| 190 |
+
"loss": 1.2847,
|
| 191 |
+
"step": 2600
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.07915103189493433,
|
| 195 |
+
"grad_norm": 0.32774218916893005,
|
| 196 |
+
"learning_rate": 4.605594979766583e-05,
|
| 197 |
+
"loss": 1.2889,
|
| 198 |
+
"step": 2700
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.08208255159474671,
|
| 202 |
+
"grad_norm": 0.26954421401023865,
|
| 203 |
+
"learning_rate": 4.590933083103631e-05,
|
| 204 |
+
"loss": 1.2823,
|
| 205 |
+
"step": 2800
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.0850140712945591,
|
| 209 |
+
"grad_norm": 0.4042912423610687,
|
| 210 |
+
"learning_rate": 4.5762711864406784e-05,
|
| 211 |
+
"loss": 1.2869,
|
| 212 |
+
"step": 2900
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.08794559099437148,
|
| 216 |
+
"grad_norm": 0.28043991327285767,
|
| 217 |
+
"learning_rate": 4.561609289777726e-05,
|
| 218 |
+
"loss": 1.2851,
|
| 219 |
+
"step": 3000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.09087711069418386,
|
| 223 |
+
"grad_norm": 0.35115665197372437,
|
| 224 |
+
"learning_rate": 4.546947393114774e-05,
|
| 225 |
+
"loss": 1.2816,
|
| 226 |
+
"step": 3100
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.09380863039399624,
|
| 230 |
+
"grad_norm": 0.3325822651386261,
|
| 231 |
+
"learning_rate": 4.532285496451821e-05,
|
| 232 |
+
"loss": 1.2821,
|
| 233 |
+
"step": 3200
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.09674015009380862,
|
| 237 |
+
"grad_norm": 0.14176355302333832,
|
| 238 |
+
"learning_rate": 4.517623599788869e-05,
|
| 239 |
+
"loss": 1.2808,
|
| 240 |
+
"step": 3300
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.099671669793621,
|
| 244 |
+
"grad_norm": 0.3418448567390442,
|
| 245 |
+
"learning_rate": 4.502961703125916e-05,
|
| 246 |
+
"loss": 1.2808,
|
| 247 |
+
"step": 3400
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.1026031894934334,
|
| 251 |
+
"grad_norm": 0.3579727113246918,
|
| 252 |
+
"learning_rate": 4.488299806462964e-05,
|
| 253 |
+
"loss": 1.2768,
|
| 254 |
+
"step": 3500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.10553470919324578,
|
| 258 |
+
"grad_norm": 0.2470945566892624,
|
| 259 |
+
"learning_rate": 4.473637909800012e-05,
|
| 260 |
+
"loss": 1.2812,
|
| 261 |
+
"step": 3600
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.10846622889305817,
|
| 265 |
+
"grad_norm": 0.32134631276130676,
|
| 266 |
+
"learning_rate": 4.45897601313706e-05,
|
| 267 |
+
"loss": 1.2797,
|
| 268 |
+
"step": 3700
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.11139774859287055,
|
| 272 |
+
"grad_norm": 0.2982274889945984,
|
| 273 |
+
"learning_rate": 4.444314116474108e-05,
|
| 274 |
+
"loss": 1.2762,
|
| 275 |
+
"step": 3800
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.11432926829268293,
|
| 279 |
+
"grad_norm": 0.2519618570804596,
|
| 280 |
+
"learning_rate": 4.429652219811155e-05,
|
| 281 |
+
"loss": 1.2767,
|
| 282 |
+
"step": 3900
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.11726078799249531,
|
| 286 |
+
"grad_norm": 0.30841368436813354,
|
| 287 |
+
"learning_rate": 4.4149903231482026e-05,
|
| 288 |
+
"loss": 1.2765,
|
| 289 |
+
"step": 4000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.1201923076923077,
|
| 293 |
+
"grad_norm": 0.3413415551185608,
|
| 294 |
+
"learning_rate": 4.40032842648525e-05,
|
| 295 |
+
"loss": 1.274,
|
| 296 |
+
"step": 4100
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.12312382739212008,
|
| 300 |
+
"grad_norm": 0.3912579119205475,
|
| 301 |
+
"learning_rate": 4.385666529822298e-05,
|
| 302 |
+
"loss": 1.2694,
|
| 303 |
+
"step": 4200
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.12605534709193245,
|
| 307 |
+
"grad_norm": 0.39514127373695374,
|
| 308 |
+
"learning_rate": 4.3710046331593455e-05,
|
| 309 |
+
"loss": 1.2742,
|
| 310 |
+
"step": 4300
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.12898686679174484,
|
| 314 |
+
"grad_norm": 0.5046316385269165,
|
| 315 |
+
"learning_rate": 4.356342736496393e-05,
|
| 316 |
+
"loss": 1.2723,
|
| 317 |
+
"step": 4400
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.1319183864915572,
|
| 321 |
+
"grad_norm": 0.40433749556541443,
|
| 322 |
+
"learning_rate": 4.341680839833441e-05,
|
| 323 |
+
"loss": 1.2714,
|
| 324 |
+
"step": 4500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.1348499061913696,
|
| 328 |
+
"grad_norm": 0.41701552271842957,
|
| 329 |
+
"learning_rate": 4.327018943170489e-05,
|
| 330 |
+
"loss": 1.2686,
|
| 331 |
+
"step": 4600
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.137781425891182,
|
| 335 |
+
"grad_norm": 0.6152161359786987,
|
| 336 |
+
"learning_rate": 4.3123570465075365e-05,
|
| 337 |
+
"loss": 1.2642,
|
| 338 |
+
"step": 4700
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.14071294559099437,
|
| 342 |
+
"grad_norm": 0.6579223275184631,
|
| 343 |
+
"learning_rate": 4.297695149844584e-05,
|
| 344 |
+
"loss": 1.2648,
|
| 345 |
+
"step": 4800
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.14364446529080677,
|
| 349 |
+
"grad_norm": 0.47039562463760376,
|
| 350 |
+
"learning_rate": 4.283033253181632e-05,
|
| 351 |
+
"loss": 1.2623,
|
| 352 |
+
"step": 4900
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.14657598499061913,
|
| 356 |
+
"grad_norm": 0.5053458213806152,
|
| 357 |
+
"learning_rate": 4.2683713565186795e-05,
|
| 358 |
+
"loss": 1.2581,
|
| 359 |
+
"step": 5000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.14950750469043153,
|
| 363 |
+
"grad_norm": 0.6096323728561401,
|
| 364 |
+
"learning_rate": 4.253709459855727e-05,
|
| 365 |
+
"loss": 1.2526,
|
| 366 |
+
"step": 5100
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.1524390243902439,
|
| 370 |
+
"grad_norm": 0.6972830891609192,
|
| 371 |
+
"learning_rate": 4.239047563192775e-05,
|
| 372 |
+
"loss": 1.2541,
|
| 373 |
+
"step": 5200
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.1553705440900563,
|
| 377 |
+
"grad_norm": 0.27344340085983276,
|
| 378 |
+
"learning_rate": 4.2243856665298224e-05,
|
| 379 |
+
"loss": 1.2496,
|
| 380 |
+
"step": 5300
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.15830206378986866,
|
| 384 |
+
"grad_norm": 0.6050379276275635,
|
| 385 |
+
"learning_rate": 4.20972376986687e-05,
|
| 386 |
+
"loss": 1.243,
|
| 387 |
+
"step": 5400
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.16123358348968106,
|
| 391 |
+
"grad_norm": 0.4654715061187744,
|
| 392 |
+
"learning_rate": 4.195061873203918e-05,
|
| 393 |
+
"loss": 1.2398,
|
| 394 |
+
"step": 5500
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.16416510318949343,
|
| 398 |
+
"grad_norm": 0.8257409334182739,
|
| 399 |
+
"learning_rate": 4.180399976540966e-05,
|
| 400 |
+
"loss": 1.2377,
|
| 401 |
+
"step": 5600
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.16709662288930582,
|
| 405 |
+
"grad_norm": 0.74887615442276,
|
| 406 |
+
"learning_rate": 4.1657380798780134e-05,
|
| 407 |
+
"loss": 1.236,
|
| 408 |
+
"step": 5700
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.1700281425891182,
|
| 412 |
+
"grad_norm": 0.7653385996818542,
|
| 413 |
+
"learning_rate": 4.151076183215061e-05,
|
| 414 |
+
"loss": 1.2276,
|
| 415 |
+
"step": 5800
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.1729596622889306,
|
| 419 |
+
"grad_norm": 0.7907219529151917,
|
| 420 |
+
"learning_rate": 4.136414286552109e-05,
|
| 421 |
+
"loss": 1.221,
|
| 422 |
+
"step": 5900
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.17589118198874296,
|
| 426 |
+
"grad_norm": 0.6932777762413025,
|
| 427 |
+
"learning_rate": 4.121752389889156e-05,
|
| 428 |
+
"loss": 1.2205,
|
| 429 |
+
"step": 6000
|
| 430 |
+
}
|
| 431 |
+
],
|
| 432 |
+
"logging_steps": 100,
|
| 433 |
+
"max_steps": 34112,
|
| 434 |
+
"num_input_tokens_seen": 0,
|
| 435 |
+
"num_train_epochs": 1,
|
| 436 |
+
"save_steps": 500,
|
| 437 |
+
"stateful_callbacks": {
|
| 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": false
|
| 445 |
+
},
|
| 446 |
+
"attributes": {}
|
| 447 |
+
}
|
| 448 |
+
},
|
| 449 |
+
"total_flos": 5.0216086536192e+16,
|
| 450 |
+
"train_batch_size": 36,
|
| 451 |
+
"trial_name": null,
|
| 452 |
+
"trial_params": null
|
| 453 |
+
}
|
checkpoint-6000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9591953a84ca47c12016b5607aa6af8b03d94b1b1c3e735d4e7c3645542c9a3c
|
| 3 |
+
size 5176
|