Training in progress, step 21000, checkpoint
Browse files- checkpoint-21000/config.json +30 -0
- checkpoint-21000/generation_config.json +6 -0
- checkpoint-21000/model.safetensors +3 -0
- checkpoint-21000/optimizer.pt +3 -0
- checkpoint-21000/rng_state.pth +3 -0
- checkpoint-21000/scheduler.pt +3 -0
- checkpoint-21000/special_tokens_map.json +24 -0
- checkpoint-21000/tokenizer.json +96 -0
- checkpoint-21000/tokenizer_config.json +38 -0
- checkpoint-21000/trainer_state.json +1503 -0
- checkpoint-21000/training_args.bin +3 -0
checkpoint-21000/config.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "EleutherAI/pythia-410m",
|
| 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": 1024,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 4096,
|
| 16 |
+
"layer_norm_eps": 1e-05,
|
| 17 |
+
"max_position_embeddings": 2048,
|
| 18 |
+
"model_type": "gpt_neox",
|
| 19 |
+
"num_attention_heads": 16,
|
| 20 |
+
"num_hidden_layers": 24,
|
| 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-21000/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-21000/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ebf83157de8040d40825e8792c1477473b56e77a1a5ca7a16e59f1777819320d
|
| 3 |
+
size 1209345104
|
checkpoint-21000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e09081a62ec1124fab6af7555b8d8a67d1477d4cd4a473c1776f53e6650560a2
|
| 3 |
+
size 2418866362
|
checkpoint-21000/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-21000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9258e98d3be777b96002cb8b8197105e82ba09c96865285b4402dbd4685a01ee
|
| 3 |
+
size 1064
|
checkpoint-21000/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-21000/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-21000/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-21000/trainer_state.json
ADDED
|
@@ -0,0 +1,1503 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.06840324035921473,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 21000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0003257297159962606,
|
| 13 |
+
"grad_norm": 2.2308592796325684,
|
| 14 |
+
"learning_rate": 4.99853416853153e-05,
|
| 15 |
+
"loss": 1.4483,
|
| 16 |
+
"step": 100
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0006514594319925212,
|
| 20 |
+
"grad_norm": 2.3997225761413574,
|
| 21 |
+
"learning_rate": 4.996905466899897e-05,
|
| 22 |
+
"loss": 1.3276,
|
| 23 |
+
"step": 200
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.0009771891479887819,
|
| 27 |
+
"grad_norm": 1.4687339067459106,
|
| 28 |
+
"learning_rate": 4.995276765268264e-05,
|
| 29 |
+
"loss": 1.3394,
|
| 30 |
+
"step": 300
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.0013029188639850425,
|
| 34 |
+
"grad_norm": 0.6583470702171326,
|
| 35 |
+
"learning_rate": 4.993648063636631e-05,
|
| 36 |
+
"loss": 1.3245,
|
| 37 |
+
"step": 400
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.0016286485799813031,
|
| 41 |
+
"grad_norm": 1.6252340078353882,
|
| 42 |
+
"learning_rate": 4.992019362004997e-05,
|
| 43 |
+
"loss": 1.3249,
|
| 44 |
+
"step": 500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.0019543782959775637,
|
| 48 |
+
"grad_norm": 2.0806777477264404,
|
| 49 |
+
"learning_rate": 4.9903906603733634e-05,
|
| 50 |
+
"loss": 1.32,
|
| 51 |
+
"step": 600
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.002280108011973824,
|
| 55 |
+
"grad_norm": 1.376539707183838,
|
| 56 |
+
"learning_rate": 4.988761958741731e-05,
|
| 57 |
+
"loss": 1.3133,
|
| 58 |
+
"step": 700
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.002605837727970085,
|
| 62 |
+
"grad_norm": 2.234644889831543,
|
| 63 |
+
"learning_rate": 4.987133257110097e-05,
|
| 64 |
+
"loss": 1.3179,
|
| 65 |
+
"step": 800
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.0029315674439663454,
|
| 69 |
+
"grad_norm": 1.4599684476852417,
|
| 70 |
+
"learning_rate": 4.985504555478464e-05,
|
| 71 |
+
"loss": 1.3097,
|
| 72 |
+
"step": 900
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.0032572971599626062,
|
| 76 |
+
"grad_norm": 1.7078094482421875,
|
| 77 |
+
"learning_rate": 4.9838758538468304e-05,
|
| 78 |
+
"loss": 1.3083,
|
| 79 |
+
"step": 1000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.0035830268759588666,
|
| 83 |
+
"grad_norm": 0.6953567266464233,
|
| 84 |
+
"learning_rate": 4.9822471522151976e-05,
|
| 85 |
+
"loss": 1.3075,
|
| 86 |
+
"step": 1100
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.0039087565919551275,
|
| 90 |
+
"grad_norm": 1.225602626800537,
|
| 91 |
+
"learning_rate": 4.980618450583564e-05,
|
| 92 |
+
"loss": 1.3054,
|
| 93 |
+
"step": 1200
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.004234486307951388,
|
| 97 |
+
"grad_norm": 1.3010519742965698,
|
| 98 |
+
"learning_rate": 4.978989748951931e-05,
|
| 99 |
+
"loss": 1.3066,
|
| 100 |
+
"step": 1300
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.004560216023947648,
|
| 104 |
+
"grad_norm": 0.6475724577903748,
|
| 105 |
+
"learning_rate": 4.9773610473202974e-05,
|
| 106 |
+
"loss": 1.3109,
|
| 107 |
+
"step": 1400
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.004885945739943909,
|
| 111 |
+
"grad_norm": 1.046614646911621,
|
| 112 |
+
"learning_rate": 4.975732345688664e-05,
|
| 113 |
+
"loss": 1.3074,
|
| 114 |
+
"step": 1500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.00521167545594017,
|
| 118 |
+
"grad_norm": 1.113573670387268,
|
| 119 |
+
"learning_rate": 4.974103644057031e-05,
|
| 120 |
+
"loss": 1.3083,
|
| 121 |
+
"step": 1600
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.005537405171936431,
|
| 125 |
+
"grad_norm": 1.4273550510406494,
|
| 126 |
+
"learning_rate": 4.972474942425398e-05,
|
| 127 |
+
"loss": 1.3018,
|
| 128 |
+
"step": 1700
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.005863134887932691,
|
| 132 |
+
"grad_norm": 0.5519908666610718,
|
| 133 |
+
"learning_rate": 4.970846240793764e-05,
|
| 134 |
+
"loss": 1.2945,
|
| 135 |
+
"step": 1800
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.006188864603928952,
|
| 139 |
+
"grad_norm": 0.6653416156768799,
|
| 140 |
+
"learning_rate": 4.969217539162131e-05,
|
| 141 |
+
"loss": 1.3004,
|
| 142 |
+
"step": 1900
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.0065145943199252125,
|
| 146 |
+
"grad_norm": 0.732170581817627,
|
| 147 |
+
"learning_rate": 4.9675888375304975e-05,
|
| 148 |
+
"loss": 1.3014,
|
| 149 |
+
"step": 2000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.006840324035921473,
|
| 153 |
+
"grad_norm": 0.405608594417572,
|
| 154 |
+
"learning_rate": 4.965960135898865e-05,
|
| 155 |
+
"loss": 1.2939,
|
| 156 |
+
"step": 2100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.007166053751917733,
|
| 160 |
+
"grad_norm": 0.9849847555160522,
|
| 161 |
+
"learning_rate": 4.9643314342672306e-05,
|
| 162 |
+
"loss": 1.2922,
|
| 163 |
+
"step": 2200
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.007491783467913994,
|
| 167 |
+
"grad_norm": 0.7152832746505737,
|
| 168 |
+
"learning_rate": 4.962702732635598e-05,
|
| 169 |
+
"loss": 1.2905,
|
| 170 |
+
"step": 2300
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.007817513183910255,
|
| 174 |
+
"grad_norm": 1.1164734363555908,
|
| 175 |
+
"learning_rate": 4.9610740310039644e-05,
|
| 176 |
+
"loss": 1.3024,
|
| 177 |
+
"step": 2400
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.008143242899906516,
|
| 181 |
+
"grad_norm": 0.574243426322937,
|
| 182 |
+
"learning_rate": 4.959445329372332e-05,
|
| 183 |
+
"loss": 1.2944,
|
| 184 |
+
"step": 2500
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.008468972615902777,
|
| 188 |
+
"grad_norm": 0.6976324319839478,
|
| 189 |
+
"learning_rate": 4.9578166277406976e-05,
|
| 190 |
+
"loss": 1.2939,
|
| 191 |
+
"step": 2600
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.008794702331899037,
|
| 195 |
+
"grad_norm": 0.4648737609386444,
|
| 196 |
+
"learning_rate": 4.956187926109064e-05,
|
| 197 |
+
"loss": 1.2841,
|
| 198 |
+
"step": 2700
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.009120432047895297,
|
| 202 |
+
"grad_norm": 1.189271092414856,
|
| 203 |
+
"learning_rate": 4.9545592244774314e-05,
|
| 204 |
+
"loss": 1.294,
|
| 205 |
+
"step": 2800
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.009446161763891557,
|
| 209 |
+
"grad_norm": 0.6437670588493347,
|
| 210 |
+
"learning_rate": 4.952930522845798e-05,
|
| 211 |
+
"loss": 1.2882,
|
| 212 |
+
"step": 2900
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.009771891479887818,
|
| 216 |
+
"grad_norm": 1.591304898262024,
|
| 217 |
+
"learning_rate": 4.9513018212141646e-05,
|
| 218 |
+
"loss": 1.2805,
|
| 219 |
+
"step": 3000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.010097621195884079,
|
| 223 |
+
"grad_norm": 0.2836475670337677,
|
| 224 |
+
"learning_rate": 4.949673119582531e-05,
|
| 225 |
+
"loss": 1.2802,
|
| 226 |
+
"step": 3100
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.01042335091188034,
|
| 230 |
+
"grad_norm": 1.304417610168457,
|
| 231 |
+
"learning_rate": 4.9480444179508984e-05,
|
| 232 |
+
"loss": 1.2833,
|
| 233 |
+
"step": 3200
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.0107490806278766,
|
| 237 |
+
"grad_norm": 0.27579864859580994,
|
| 238 |
+
"learning_rate": 4.946415716319265e-05,
|
| 239 |
+
"loss": 1.2852,
|
| 240 |
+
"step": 3300
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.011074810343872862,
|
| 244 |
+
"grad_norm": 1.1080585718154907,
|
| 245 |
+
"learning_rate": 4.9447870146876315e-05,
|
| 246 |
+
"loss": 1.289,
|
| 247 |
+
"step": 3400
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.011400540059869122,
|
| 251 |
+
"grad_norm": 0.2783690392971039,
|
| 252 |
+
"learning_rate": 4.943158313055998e-05,
|
| 253 |
+
"loss": 1.2885,
|
| 254 |
+
"step": 3500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.011726269775865382,
|
| 258 |
+
"grad_norm": 0.6603112816810608,
|
| 259 |
+
"learning_rate": 4.941529611424365e-05,
|
| 260 |
+
"loss": 1.2882,
|
| 261 |
+
"step": 3600
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.012051999491861642,
|
| 265 |
+
"grad_norm": 0.9498095512390137,
|
| 266 |
+
"learning_rate": 4.939900909792732e-05,
|
| 267 |
+
"loss": 1.2835,
|
| 268 |
+
"step": 3700
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.012377729207857903,
|
| 272 |
+
"grad_norm": 0.5274548530578613,
|
| 273 |
+
"learning_rate": 4.9382722081610985e-05,
|
| 274 |
+
"loss": 1.279,
|
| 275 |
+
"step": 3800
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.012703458923854164,
|
| 279 |
+
"grad_norm": 0.5299821496009827,
|
| 280 |
+
"learning_rate": 4.936643506529465e-05,
|
| 281 |
+
"loss": 1.2879,
|
| 282 |
+
"step": 3900
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.013029188639850425,
|
| 286 |
+
"grad_norm": 1.0898863077163696,
|
| 287 |
+
"learning_rate": 4.9350148048978316e-05,
|
| 288 |
+
"loss": 1.2913,
|
| 289 |
+
"step": 4000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.013354918355846686,
|
| 293 |
+
"grad_norm": 0.6892501711845398,
|
| 294 |
+
"learning_rate": 4.933386103266198e-05,
|
| 295 |
+
"loss": 1.2835,
|
| 296 |
+
"step": 4100
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.013680648071842947,
|
| 300 |
+
"grad_norm": 0.9103847146034241,
|
| 301 |
+
"learning_rate": 4.9317574016345655e-05,
|
| 302 |
+
"loss": 1.2876,
|
| 303 |
+
"step": 4200
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.014006377787839207,
|
| 307 |
+
"grad_norm": 0.8750960826873779,
|
| 308 |
+
"learning_rate": 4.9301287000029314e-05,
|
| 309 |
+
"loss": 1.2761,
|
| 310 |
+
"step": 4300
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.014332107503835467,
|
| 314 |
+
"grad_norm": 1.7296843528747559,
|
| 315 |
+
"learning_rate": 4.9284999983712986e-05,
|
| 316 |
+
"loss": 1.2825,
|
| 317 |
+
"step": 4400
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.014657837219831727,
|
| 321 |
+
"grad_norm": 0.7019387483596802,
|
| 322 |
+
"learning_rate": 4.926871296739665e-05,
|
| 323 |
+
"loss": 1.2774,
|
| 324 |
+
"step": 4500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.014983566935827988,
|
| 328 |
+
"grad_norm": 0.9353660345077515,
|
| 329 |
+
"learning_rate": 4.9252425951080324e-05,
|
| 330 |
+
"loss": 1.2701,
|
| 331 |
+
"step": 4600
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.015309296651824249,
|
| 335 |
+
"grad_norm": 0.7081932425498962,
|
| 336 |
+
"learning_rate": 4.923613893476399e-05,
|
| 337 |
+
"loss": 1.276,
|
| 338 |
+
"step": 4700
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.01563502636782051,
|
| 342 |
+
"grad_norm": 0.8366962671279907,
|
| 343 |
+
"learning_rate": 4.9219851918447656e-05,
|
| 344 |
+
"loss": 1.2767,
|
| 345 |
+
"step": 4800
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.01596075608381677,
|
| 349 |
+
"grad_norm": 1.765871286392212,
|
| 350 |
+
"learning_rate": 4.920356490213132e-05,
|
| 351 |
+
"loss": 1.2617,
|
| 352 |
+
"step": 4900
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.01628648579981303,
|
| 356 |
+
"grad_norm": 0.2926379442214966,
|
| 357 |
+
"learning_rate": 4.918727788581499e-05,
|
| 358 |
+
"loss": 1.2762,
|
| 359 |
+
"step": 5000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.01661221551580929,
|
| 363 |
+
"grad_norm": 1.1176525354385376,
|
| 364 |
+
"learning_rate": 4.917099086949866e-05,
|
| 365 |
+
"loss": 1.2647,
|
| 366 |
+
"step": 5100
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.016937945231805553,
|
| 370 |
+
"grad_norm": 0.384264200925827,
|
| 371 |
+
"learning_rate": 4.915470385318232e-05,
|
| 372 |
+
"loss": 1.2628,
|
| 373 |
+
"step": 5200
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.017263674947801812,
|
| 377 |
+
"grad_norm": 1.5339140892028809,
|
| 378 |
+
"learning_rate": 4.913841683686599e-05,
|
| 379 |
+
"loss": 1.2692,
|
| 380 |
+
"step": 5300
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.017589404663798075,
|
| 384 |
+
"grad_norm": 1.2026703357696533,
|
| 385 |
+
"learning_rate": 4.912212982054966e-05,
|
| 386 |
+
"loss": 1.2618,
|
| 387 |
+
"step": 5400
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.017915134379794334,
|
| 391 |
+
"grad_norm": 0.6754997968673706,
|
| 392 |
+
"learning_rate": 4.910584280423333e-05,
|
| 393 |
+
"loss": 1.2495,
|
| 394 |
+
"step": 5500
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.018240864095790593,
|
| 398 |
+
"grad_norm": 0.8240428566932678,
|
| 399 |
+
"learning_rate": 4.908955578791699e-05,
|
| 400 |
+
"loss": 1.2498,
|
| 401 |
+
"step": 5600
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.018566593811786856,
|
| 405 |
+
"grad_norm": 0.6363087892532349,
|
| 406 |
+
"learning_rate": 4.9073268771600654e-05,
|
| 407 |
+
"loss": 1.2514,
|
| 408 |
+
"step": 5700
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.018892323527783115,
|
| 412 |
+
"grad_norm": 1.393833875656128,
|
| 413 |
+
"learning_rate": 4.905698175528433e-05,
|
| 414 |
+
"loss": 1.2509,
|
| 415 |
+
"step": 5800
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.019218053243779377,
|
| 419 |
+
"grad_norm": 0.6422170996665955,
|
| 420 |
+
"learning_rate": 4.904069473896799e-05,
|
| 421 |
+
"loss": 1.2405,
|
| 422 |
+
"step": 5900
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.019543782959775637,
|
| 426 |
+
"grad_norm": 0.7575420141220093,
|
| 427 |
+
"learning_rate": 4.902440772265166e-05,
|
| 428 |
+
"loss": 1.2241,
|
| 429 |
+
"step": 6000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.0198695126757719,
|
| 433 |
+
"grad_norm": 0.7148196697235107,
|
| 434 |
+
"learning_rate": 4.9008120706335324e-05,
|
| 435 |
+
"loss": 1.2372,
|
| 436 |
+
"step": 6100
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.020195242391768158,
|
| 440 |
+
"grad_norm": 1.1207329034805298,
|
| 441 |
+
"learning_rate": 4.8991833690018996e-05,
|
| 442 |
+
"loss": 1.2372,
|
| 443 |
+
"step": 6200
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.02052097210776442,
|
| 447 |
+
"grad_norm": 1.3915568590164185,
|
| 448 |
+
"learning_rate": 4.897554667370266e-05,
|
| 449 |
+
"loss": 1.2129,
|
| 450 |
+
"step": 6300
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.02084670182376068,
|
| 454 |
+
"grad_norm": 0.8674553036689758,
|
| 455 |
+
"learning_rate": 4.895925965738633e-05,
|
| 456 |
+
"loss": 1.2262,
|
| 457 |
+
"step": 6400
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.02117243153975694,
|
| 461 |
+
"grad_norm": 0.7640644311904907,
|
| 462 |
+
"learning_rate": 4.8942972641069994e-05,
|
| 463 |
+
"loss": 1.1998,
|
| 464 |
+
"step": 6500
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.0214981612557532,
|
| 468 |
+
"grad_norm": 0.7928606271743774,
|
| 469 |
+
"learning_rate": 4.892668562475366e-05,
|
| 470 |
+
"loss": 1.1776,
|
| 471 |
+
"step": 6600
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.02182389097174946,
|
| 475 |
+
"grad_norm": 1.1644946336746216,
|
| 476 |
+
"learning_rate": 4.891039860843733e-05,
|
| 477 |
+
"loss": 1.1916,
|
| 478 |
+
"step": 6700
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.022149620687745723,
|
| 482 |
+
"grad_norm": 1.1310213804244995,
|
| 483 |
+
"learning_rate": 4.8894111592121e-05,
|
| 484 |
+
"loss": 1.1786,
|
| 485 |
+
"step": 6800
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.022475350403741982,
|
| 489 |
+
"grad_norm": 1.3858141899108887,
|
| 490 |
+
"learning_rate": 4.887782457580466e-05,
|
| 491 |
+
"loss": 1.1728,
|
| 492 |
+
"step": 6900
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.022801080119738245,
|
| 496 |
+
"grad_norm": 3.814767360687256,
|
| 497 |
+
"learning_rate": 4.886153755948833e-05,
|
| 498 |
+
"loss": 1.1384,
|
| 499 |
+
"step": 7000
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.023126809835734504,
|
| 503 |
+
"grad_norm": 1.2411885261535645,
|
| 504 |
+
"learning_rate": 4.8845250543171995e-05,
|
| 505 |
+
"loss": 1.1588,
|
| 506 |
+
"step": 7100
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.023452539551730763,
|
| 510 |
+
"grad_norm": 1.4492881298065186,
|
| 511 |
+
"learning_rate": 4.882896352685567e-05,
|
| 512 |
+
"loss": 1.1266,
|
| 513 |
+
"step": 7200
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.023778269267727026,
|
| 517 |
+
"grad_norm": 0.8389878869056702,
|
| 518 |
+
"learning_rate": 4.8812676510539326e-05,
|
| 519 |
+
"loss": 1.1446,
|
| 520 |
+
"step": 7300
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.024103998983723285,
|
| 524 |
+
"grad_norm": 0.33955487608909607,
|
| 525 |
+
"learning_rate": 4.8796389494223e-05,
|
| 526 |
+
"loss": 1.1111,
|
| 527 |
+
"step": 7400
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.024429728699719547,
|
| 531 |
+
"grad_norm": 0.7004753351211548,
|
| 532 |
+
"learning_rate": 4.8780102477906664e-05,
|
| 533 |
+
"loss": 1.0954,
|
| 534 |
+
"step": 7500
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.024755458415715807,
|
| 538 |
+
"grad_norm": 0.7213209271430969,
|
| 539 |
+
"learning_rate": 4.876381546159034e-05,
|
| 540 |
+
"loss": 1.1123,
|
| 541 |
+
"step": 7600
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.02508118813171207,
|
| 545 |
+
"grad_norm": 0.960991382598877,
|
| 546 |
+
"learning_rate": 4.8747528445273996e-05,
|
| 547 |
+
"loss": 1.0982,
|
| 548 |
+
"step": 7700
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.025406917847708328,
|
| 552 |
+
"grad_norm": 0.6955804228782654,
|
| 553 |
+
"learning_rate": 4.873124142895766e-05,
|
| 554 |
+
"loss": 1.0827,
|
| 555 |
+
"step": 7800
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.02573264756370459,
|
| 559 |
+
"grad_norm": 0.47498619556427,
|
| 560 |
+
"learning_rate": 4.8714954412641334e-05,
|
| 561 |
+
"loss": 1.1043,
|
| 562 |
+
"step": 7900
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.02605837727970085,
|
| 566 |
+
"grad_norm": 0.304063618183136,
|
| 567 |
+
"learning_rate": 4.8698667396325e-05,
|
| 568 |
+
"loss": 1.0699,
|
| 569 |
+
"step": 8000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.02638410699569711,
|
| 573 |
+
"grad_norm": 0.9996088743209839,
|
| 574 |
+
"learning_rate": 4.8682380380008666e-05,
|
| 575 |
+
"loss": 1.0697,
|
| 576 |
+
"step": 8100
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.02670983671169337,
|
| 580 |
+
"grad_norm": 0.5986392498016357,
|
| 581 |
+
"learning_rate": 4.866609336369233e-05,
|
| 582 |
+
"loss": 1.0733,
|
| 583 |
+
"step": 8200
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.02703556642768963,
|
| 587 |
+
"grad_norm": 0.41347017884254456,
|
| 588 |
+
"learning_rate": 4.8649806347376004e-05,
|
| 589 |
+
"loss": 1.0643,
|
| 590 |
+
"step": 8300
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.027361296143685893,
|
| 594 |
+
"grad_norm": 0.3976612687110901,
|
| 595 |
+
"learning_rate": 4.863351933105967e-05,
|
| 596 |
+
"loss": 1.0401,
|
| 597 |
+
"step": 8400
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.027687025859682152,
|
| 601 |
+
"grad_norm": 1.1716387271881104,
|
| 602 |
+
"learning_rate": 4.8617232314743335e-05,
|
| 603 |
+
"loss": 1.0298,
|
| 604 |
+
"step": 8500
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.028012755575678415,
|
| 608 |
+
"grad_norm": 0.7384105324745178,
|
| 609 |
+
"learning_rate": 4.8600945298427e-05,
|
| 610 |
+
"loss": 1.0223,
|
| 611 |
+
"step": 8600
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.028338485291674674,
|
| 615 |
+
"grad_norm": 0.517280638217926,
|
| 616 |
+
"learning_rate": 4.858465828211067e-05,
|
| 617 |
+
"loss": 1.0445,
|
| 618 |
+
"step": 8700
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.028664215007670933,
|
| 622 |
+
"grad_norm": 0.7129126787185669,
|
| 623 |
+
"learning_rate": 4.856837126579434e-05,
|
| 624 |
+
"loss": 1.0508,
|
| 625 |
+
"step": 8800
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.028989944723667196,
|
| 629 |
+
"grad_norm": 0.35596320033073425,
|
| 630 |
+
"learning_rate": 4.8552084249478005e-05,
|
| 631 |
+
"loss": 1.0296,
|
| 632 |
+
"step": 8900
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.029315674439663455,
|
| 636 |
+
"grad_norm": 0.9362590909004211,
|
| 637 |
+
"learning_rate": 4.853579723316167e-05,
|
| 638 |
+
"loss": 1.0785,
|
| 639 |
+
"step": 9000
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.029641404155659717,
|
| 643 |
+
"grad_norm": 0.8223775625228882,
|
| 644 |
+
"learning_rate": 4.8519510216845336e-05,
|
| 645 |
+
"loss": 1.043,
|
| 646 |
+
"step": 9100
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.029967133871655977,
|
| 650 |
+
"grad_norm": 0.7149192690849304,
|
| 651 |
+
"learning_rate": 4.8503223200529e-05,
|
| 652 |
+
"loss": 1.0036,
|
| 653 |
+
"step": 9200
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.03029286358765224,
|
| 657 |
+
"grad_norm": 0.5907948017120361,
|
| 658 |
+
"learning_rate": 4.8486936184212675e-05,
|
| 659 |
+
"loss": 1.0408,
|
| 660 |
+
"step": 9300
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.030618593303648498,
|
| 664 |
+
"grad_norm": 0.6083859801292419,
|
| 665 |
+
"learning_rate": 4.847064916789634e-05,
|
| 666 |
+
"loss": 1.0313,
|
| 667 |
+
"step": 9400
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.03094432301964476,
|
| 671 |
+
"grad_norm": 0.5470224618911743,
|
| 672 |
+
"learning_rate": 4.8454362151580006e-05,
|
| 673 |
+
"loss": 1.0395,
|
| 674 |
+
"step": 9500
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.03127005273564102,
|
| 678 |
+
"grad_norm": 0.9455150961875916,
|
| 679 |
+
"learning_rate": 4.843807513526367e-05,
|
| 680 |
+
"loss": 1.0132,
|
| 681 |
+
"step": 9600
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.03159578245163728,
|
| 685 |
+
"grad_norm": 0.9068177938461304,
|
| 686 |
+
"learning_rate": 4.8421788118947344e-05,
|
| 687 |
+
"loss": 1.0219,
|
| 688 |
+
"step": 9700
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.03192151216763354,
|
| 692 |
+
"grad_norm": 0.6018943190574646,
|
| 693 |
+
"learning_rate": 4.840550110263101e-05,
|
| 694 |
+
"loss": 0.9966,
|
| 695 |
+
"step": 9800
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.032247241883629804,
|
| 699 |
+
"grad_norm": 1.1521615982055664,
|
| 700 |
+
"learning_rate": 4.838921408631467e-05,
|
| 701 |
+
"loss": 0.9782,
|
| 702 |
+
"step": 9900
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.03257297159962606,
|
| 706 |
+
"grad_norm": 0.33281368017196655,
|
| 707 |
+
"learning_rate": 4.837292706999834e-05,
|
| 708 |
+
"loss": 1.0325,
|
| 709 |
+
"step": 10000
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.03289870131562232,
|
| 713 |
+
"grad_norm": 0.8903327584266663,
|
| 714 |
+
"learning_rate": 4.835664005368201e-05,
|
| 715 |
+
"loss": 0.9889,
|
| 716 |
+
"step": 10100
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.03322443103161858,
|
| 720 |
+
"grad_norm": 0.5526803731918335,
|
| 721 |
+
"learning_rate": 4.834035303736568e-05,
|
| 722 |
+
"loss": 1.0018,
|
| 723 |
+
"step": 10200
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.03355016074761485,
|
| 727 |
+
"grad_norm": 0.8086706399917603,
|
| 728 |
+
"learning_rate": 4.832406602104934e-05,
|
| 729 |
+
"loss": 1.0189,
|
| 730 |
+
"step": 10300
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.03387589046361111,
|
| 734 |
+
"grad_norm": 0.6990864276885986,
|
| 735 |
+
"learning_rate": 4.830777900473301e-05,
|
| 736 |
+
"loss": 0.996,
|
| 737 |
+
"step": 10400
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.034201620179607366,
|
| 741 |
+
"grad_norm": 0.4859602451324463,
|
| 742 |
+
"learning_rate": 4.829149198841668e-05,
|
| 743 |
+
"loss": 0.992,
|
| 744 |
+
"step": 10500
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.034527349895603625,
|
| 748 |
+
"grad_norm": 1.2284592390060425,
|
| 749 |
+
"learning_rate": 4.827520497210034e-05,
|
| 750 |
+
"loss": 1.0139,
|
| 751 |
+
"step": 10600
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.034853079611599884,
|
| 755 |
+
"grad_norm": 0.6529733538627625,
|
| 756 |
+
"learning_rate": 4.825891795578401e-05,
|
| 757 |
+
"loss": 1.025,
|
| 758 |
+
"step": 10700
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.03517880932759615,
|
| 762 |
+
"grad_norm": 0.6755232810974121,
|
| 763 |
+
"learning_rate": 4.8242630939467674e-05,
|
| 764 |
+
"loss": 1.0123,
|
| 765 |
+
"step": 10800
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.03550453904359241,
|
| 769 |
+
"grad_norm": 0.9006055593490601,
|
| 770 |
+
"learning_rate": 4.8226343923151347e-05,
|
| 771 |
+
"loss": 0.9936,
|
| 772 |
+
"step": 10900
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.03583026875958867,
|
| 776 |
+
"grad_norm": 0.7058572769165039,
|
| 777 |
+
"learning_rate": 4.821005690683501e-05,
|
| 778 |
+
"loss": 0.934,
|
| 779 |
+
"step": 11000
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.03615599847558493,
|
| 783 |
+
"grad_norm": 0.4535008668899536,
|
| 784 |
+
"learning_rate": 4.819376989051868e-05,
|
| 785 |
+
"loss": 1.0269,
|
| 786 |
+
"step": 11100
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.036481728191581186,
|
| 790 |
+
"grad_norm": 0.39823395013809204,
|
| 791 |
+
"learning_rate": 4.8177482874202344e-05,
|
| 792 |
+
"loss": 0.9866,
|
| 793 |
+
"step": 11200
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.03680745790757745,
|
| 797 |
+
"grad_norm": 0.8109054565429688,
|
| 798 |
+
"learning_rate": 4.816119585788601e-05,
|
| 799 |
+
"loss": 1.0209,
|
| 800 |
+
"step": 11300
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.03713318762357371,
|
| 804 |
+
"grad_norm": 0.760396420955658,
|
| 805 |
+
"learning_rate": 4.814490884156968e-05,
|
| 806 |
+
"loss": 0.9711,
|
| 807 |
+
"step": 11400
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.03745891733956997,
|
| 811 |
+
"grad_norm": 0.8584955334663391,
|
| 812 |
+
"learning_rate": 4.812862182525335e-05,
|
| 813 |
+
"loss": 1.0151,
|
| 814 |
+
"step": 11500
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.03778464705556623,
|
| 818 |
+
"grad_norm": 1.104041576385498,
|
| 819 |
+
"learning_rate": 4.8112334808937013e-05,
|
| 820 |
+
"loss": 0.9826,
|
| 821 |
+
"step": 11600
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.038110376771562496,
|
| 825 |
+
"grad_norm": 0.6111257672309875,
|
| 826 |
+
"learning_rate": 4.809604779262068e-05,
|
| 827 |
+
"loss": 0.9524,
|
| 828 |
+
"step": 11700
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.038436106487558755,
|
| 832 |
+
"grad_norm": 0.6601366996765137,
|
| 833 |
+
"learning_rate": 4.807976077630435e-05,
|
| 834 |
+
"loss": 0.9527,
|
| 835 |
+
"step": 11800
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.038761836203555014,
|
| 839 |
+
"grad_norm": 0.4624398350715637,
|
| 840 |
+
"learning_rate": 4.806347375998802e-05,
|
| 841 |
+
"loss": 1.0077,
|
| 842 |
+
"step": 11900
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.03908756591955127,
|
| 846 |
+
"grad_norm": 0.2786065638065338,
|
| 847 |
+
"learning_rate": 4.8047186743671676e-05,
|
| 848 |
+
"loss": 0.956,
|
| 849 |
+
"step": 12000
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.03941329563554753,
|
| 853 |
+
"grad_norm": 1.0275955200195312,
|
| 854 |
+
"learning_rate": 4.803089972735535e-05,
|
| 855 |
+
"loss": 0.9484,
|
| 856 |
+
"step": 12100
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.0397390253515438,
|
| 860 |
+
"grad_norm": 0.6198407411575317,
|
| 861 |
+
"learning_rate": 4.8014612711039015e-05,
|
| 862 |
+
"loss": 0.9847,
|
| 863 |
+
"step": 12200
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.04006475506754006,
|
| 867 |
+
"grad_norm": 0.5880489945411682,
|
| 868 |
+
"learning_rate": 4.799832569472269e-05,
|
| 869 |
+
"loss": 0.9559,
|
| 870 |
+
"step": 12300
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.040390484783536316,
|
| 874 |
+
"grad_norm": 0.39753594994544983,
|
| 875 |
+
"learning_rate": 4.7982038678406346e-05,
|
| 876 |
+
"loss": 0.9489,
|
| 877 |
+
"step": 12400
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.040716214499532576,
|
| 881 |
+
"grad_norm": 0.5815085768699646,
|
| 882 |
+
"learning_rate": 4.796575166209002e-05,
|
| 883 |
+
"loss": 0.9567,
|
| 884 |
+
"step": 12500
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.04104194421552884,
|
| 888 |
+
"grad_norm": 0.8463611602783203,
|
| 889 |
+
"learning_rate": 4.7949464645773684e-05,
|
| 890 |
+
"loss": 0.9706,
|
| 891 |
+
"step": 12600
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.0413676739315251,
|
| 895 |
+
"grad_norm": 0.7260481715202332,
|
| 896 |
+
"learning_rate": 4.793317762945736e-05,
|
| 897 |
+
"loss": 1.0032,
|
| 898 |
+
"step": 12700
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.04169340364752136,
|
| 902 |
+
"grad_norm": 0.6970434188842773,
|
| 903 |
+
"learning_rate": 4.7916890613141016e-05,
|
| 904 |
+
"loss": 0.9559,
|
| 905 |
+
"step": 12800
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.04201913336351762,
|
| 909 |
+
"grad_norm": 0.6083927750587463,
|
| 910 |
+
"learning_rate": 4.790060359682468e-05,
|
| 911 |
+
"loss": 0.9558,
|
| 912 |
+
"step": 12900
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.04234486307951388,
|
| 916 |
+
"grad_norm": 0.4736403524875641,
|
| 917 |
+
"learning_rate": 4.7884316580508354e-05,
|
| 918 |
+
"loss": 0.9444,
|
| 919 |
+
"step": 13000
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.042670592795510144,
|
| 923 |
+
"grad_norm": 0.34586021304130554,
|
| 924 |
+
"learning_rate": 4.786802956419202e-05,
|
| 925 |
+
"loss": 0.9186,
|
| 926 |
+
"step": 13100
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.0429963225115064,
|
| 930 |
+
"grad_norm": 0.5979019403457642,
|
| 931 |
+
"learning_rate": 4.7851742547875685e-05,
|
| 932 |
+
"loss": 0.9367,
|
| 933 |
+
"step": 13200
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.04332205222750266,
|
| 937 |
+
"grad_norm": 1.0827624797821045,
|
| 938 |
+
"learning_rate": 4.783545553155935e-05,
|
| 939 |
+
"loss": 0.9324,
|
| 940 |
+
"step": 13300
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.04364778194349892,
|
| 944 |
+
"grad_norm": 1.1920030117034912,
|
| 945 |
+
"learning_rate": 4.7819168515243024e-05,
|
| 946 |
+
"loss": 0.9367,
|
| 947 |
+
"step": 13400
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.04397351165949519,
|
| 951 |
+
"grad_norm": 0.6469812989234924,
|
| 952 |
+
"learning_rate": 4.780288149892669e-05,
|
| 953 |
+
"loss": 0.9815,
|
| 954 |
+
"step": 13500
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.04429924137549145,
|
| 958 |
+
"grad_norm": 0.8156530857086182,
|
| 959 |
+
"learning_rate": 4.7786594482610355e-05,
|
| 960 |
+
"loss": 0.9679,
|
| 961 |
+
"step": 13600
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.044624971091487706,
|
| 965 |
+
"grad_norm": 1.2997325658798218,
|
| 966 |
+
"learning_rate": 4.777030746629402e-05,
|
| 967 |
+
"loss": 0.9358,
|
| 968 |
+
"step": 13700
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.044950700807483965,
|
| 972 |
+
"grad_norm": 0.42360150814056396,
|
| 973 |
+
"learning_rate": 4.7754020449977687e-05,
|
| 974 |
+
"loss": 0.9326,
|
| 975 |
+
"step": 13800
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.045276430523480224,
|
| 979 |
+
"grad_norm": 0.7316247820854187,
|
| 980 |
+
"learning_rate": 4.773773343366136e-05,
|
| 981 |
+
"loss": 0.9283,
|
| 982 |
+
"step": 13900
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.04560216023947649,
|
| 986 |
+
"grad_norm": 0.5978175401687622,
|
| 987 |
+
"learning_rate": 4.7721446417345025e-05,
|
| 988 |
+
"loss": 0.9699,
|
| 989 |
+
"step": 14000
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.04592788995547275,
|
| 993 |
+
"grad_norm": 0.5278334617614746,
|
| 994 |
+
"learning_rate": 4.770515940102869e-05,
|
| 995 |
+
"loss": 0.99,
|
| 996 |
+
"step": 14100
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.04625361967146901,
|
| 1000 |
+
"grad_norm": 0.7452822327613831,
|
| 1001 |
+
"learning_rate": 4.7688872384712356e-05,
|
| 1002 |
+
"loss": 0.8824,
|
| 1003 |
+
"step": 14200
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.04657934938746527,
|
| 1007 |
+
"grad_norm": 0.4158065617084503,
|
| 1008 |
+
"learning_rate": 4.767258536839602e-05,
|
| 1009 |
+
"loss": 0.9076,
|
| 1010 |
+
"step": 14300
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.046905079103461526,
|
| 1014 |
+
"grad_norm": 0.6929590106010437,
|
| 1015 |
+
"learning_rate": 4.7656298352079694e-05,
|
| 1016 |
+
"loss": 0.926,
|
| 1017 |
+
"step": 14400
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.04723080881945779,
|
| 1021 |
+
"grad_norm": 0.8249752521514893,
|
| 1022 |
+
"learning_rate": 4.764001133576336e-05,
|
| 1023 |
+
"loss": 0.9342,
|
| 1024 |
+
"step": 14500
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.04755653853545405,
|
| 1028 |
+
"grad_norm": 0.6523115038871765,
|
| 1029 |
+
"learning_rate": 4.7623724319447026e-05,
|
| 1030 |
+
"loss": 0.9312,
|
| 1031 |
+
"step": 14600
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.04788226825145031,
|
| 1035 |
+
"grad_norm": 0.7809571027755737,
|
| 1036 |
+
"learning_rate": 4.760743730313069e-05,
|
| 1037 |
+
"loss": 0.927,
|
| 1038 |
+
"step": 14700
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.04820799796744657,
|
| 1042 |
+
"grad_norm": 0.4370424747467041,
|
| 1043 |
+
"learning_rate": 4.7591150286814364e-05,
|
| 1044 |
+
"loss": 0.9275,
|
| 1045 |
+
"step": 14800
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.048533727683442836,
|
| 1049 |
+
"grad_norm": 0.8082228302955627,
|
| 1050 |
+
"learning_rate": 4.757486327049803e-05,
|
| 1051 |
+
"loss": 0.9524,
|
| 1052 |
+
"step": 14900
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.048859457399439095,
|
| 1056 |
+
"grad_norm": 0.7073273658752441,
|
| 1057 |
+
"learning_rate": 4.755857625418169e-05,
|
| 1058 |
+
"loss": 0.9069,
|
| 1059 |
+
"step": 15000
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.049185187115435354,
|
| 1063 |
+
"grad_norm": 0.9150802493095398,
|
| 1064 |
+
"learning_rate": 4.754228923786536e-05,
|
| 1065 |
+
"loss": 0.9669,
|
| 1066 |
+
"step": 15100
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.04951091683143161,
|
| 1070 |
+
"grad_norm": 0.6621295809745789,
|
| 1071 |
+
"learning_rate": 4.752600222154903e-05,
|
| 1072 |
+
"loss": 0.9117,
|
| 1073 |
+
"step": 15200
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.04983664654742787,
|
| 1077 |
+
"grad_norm": 1.1658425331115723,
|
| 1078 |
+
"learning_rate": 4.75097152052327e-05,
|
| 1079 |
+
"loss": 0.9061,
|
| 1080 |
+
"step": 15300
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.05016237626342414,
|
| 1084 |
+
"grad_norm": 1.1669522523880005,
|
| 1085 |
+
"learning_rate": 4.749342818891636e-05,
|
| 1086 |
+
"loss": 0.9625,
|
| 1087 |
+
"step": 15400
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.0504881059794204,
|
| 1091 |
+
"grad_norm": 0.6995384693145752,
|
| 1092 |
+
"learning_rate": 4.747714117260003e-05,
|
| 1093 |
+
"loss": 0.9098,
|
| 1094 |
+
"step": 15500
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.050813835695416656,
|
| 1098 |
+
"grad_norm": 0.5169076919555664,
|
| 1099 |
+
"learning_rate": 4.74608541562837e-05,
|
| 1100 |
+
"loss": 0.9243,
|
| 1101 |
+
"step": 15600
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.051139565411412916,
|
| 1105 |
+
"grad_norm": 0.33565372228622437,
|
| 1106 |
+
"learning_rate": 4.744456713996736e-05,
|
| 1107 |
+
"loss": 0.9375,
|
| 1108 |
+
"step": 15700
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.05146529512740918,
|
| 1112 |
+
"grad_norm": 0.4140024781227112,
|
| 1113 |
+
"learning_rate": 4.742828012365103e-05,
|
| 1114 |
+
"loss": 0.919,
|
| 1115 |
+
"step": 15800
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.05179102484340544,
|
| 1119 |
+
"grad_norm": 0.9499224424362183,
|
| 1120 |
+
"learning_rate": 4.7411993107334694e-05,
|
| 1121 |
+
"loss": 0.9034,
|
| 1122 |
+
"step": 15900
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.0521167545594017,
|
| 1126 |
+
"grad_norm": 0.8801336288452148,
|
| 1127 |
+
"learning_rate": 4.7395706091018366e-05,
|
| 1128 |
+
"loss": 0.881,
|
| 1129 |
+
"step": 16000
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.05244248427539796,
|
| 1133 |
+
"grad_norm": 0.7208696007728577,
|
| 1134 |
+
"learning_rate": 4.737941907470203e-05,
|
| 1135 |
+
"loss": 0.8518,
|
| 1136 |
+
"step": 16100
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.05276821399139422,
|
| 1140 |
+
"grad_norm": 0.5132054686546326,
|
| 1141 |
+
"learning_rate": 4.73631320583857e-05,
|
| 1142 |
+
"loss": 0.8933,
|
| 1143 |
+
"step": 16200
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.053093943707390484,
|
| 1147 |
+
"grad_norm": 0.6521860957145691,
|
| 1148 |
+
"learning_rate": 4.7346845042069364e-05,
|
| 1149 |
+
"loss": 0.9332,
|
| 1150 |
+
"step": 16300
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.05341967342338674,
|
| 1154 |
+
"grad_norm": 0.7121620178222656,
|
| 1155 |
+
"learning_rate": 4.733055802575303e-05,
|
| 1156 |
+
"loss": 0.9067,
|
| 1157 |
+
"step": 16400
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.053745403139383,
|
| 1161 |
+
"grad_norm": 0.5065134763717651,
|
| 1162 |
+
"learning_rate": 4.73142710094367e-05,
|
| 1163 |
+
"loss": 0.9062,
|
| 1164 |
+
"step": 16500
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.05407113285537926,
|
| 1168 |
+
"grad_norm": 0.5855521559715271,
|
| 1169 |
+
"learning_rate": 4.729798399312037e-05,
|
| 1170 |
+
"loss": 0.915,
|
| 1171 |
+
"step": 16600
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.05439686257137553,
|
| 1175 |
+
"grad_norm": 0.5392531156539917,
|
| 1176 |
+
"learning_rate": 4.728169697680403e-05,
|
| 1177 |
+
"loss": 0.9124,
|
| 1178 |
+
"step": 16700
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.05472259228737179,
|
| 1182 |
+
"grad_norm": 0.6617989540100098,
|
| 1183 |
+
"learning_rate": 4.72654099604877e-05,
|
| 1184 |
+
"loss": 0.8594,
|
| 1185 |
+
"step": 16800
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.055048322003368046,
|
| 1189 |
+
"grad_norm": 0.6459785103797913,
|
| 1190 |
+
"learning_rate": 4.724912294417137e-05,
|
| 1191 |
+
"loss": 0.9262,
|
| 1192 |
+
"step": 16900
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.055374051719364305,
|
| 1196 |
+
"grad_norm": 0.34565970301628113,
|
| 1197 |
+
"learning_rate": 4.723283592785504e-05,
|
| 1198 |
+
"loss": 0.8747,
|
| 1199 |
+
"step": 17000
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.055699781435360564,
|
| 1203 |
+
"grad_norm": 0.9510948061943054,
|
| 1204 |
+
"learning_rate": 4.7216548911538696e-05,
|
| 1205 |
+
"loss": 0.9027,
|
| 1206 |
+
"step": 17100
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.05602551115135683,
|
| 1210 |
+
"grad_norm": 0.577192485332489,
|
| 1211 |
+
"learning_rate": 4.720026189522237e-05,
|
| 1212 |
+
"loss": 0.9192,
|
| 1213 |
+
"step": 17200
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.05635124086735309,
|
| 1217 |
+
"grad_norm": 0.38653406500816345,
|
| 1218 |
+
"learning_rate": 4.7183974878906034e-05,
|
| 1219 |
+
"loss": 0.8759,
|
| 1220 |
+
"step": 17300
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.05667697058334935,
|
| 1224 |
+
"grad_norm": 0.6405381560325623,
|
| 1225 |
+
"learning_rate": 4.716768786258971e-05,
|
| 1226 |
+
"loss": 0.8486,
|
| 1227 |
+
"step": 17400
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.05700270029934561,
|
| 1231 |
+
"grad_norm": 0.6968704462051392,
|
| 1232 |
+
"learning_rate": 4.7151400846273366e-05,
|
| 1233 |
+
"loss": 0.903,
|
| 1234 |
+
"step": 17500
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.057328430015341866,
|
| 1238 |
+
"grad_norm": 0.8094695210456848,
|
| 1239 |
+
"learning_rate": 4.713511382995704e-05,
|
| 1240 |
+
"loss": 0.864,
|
| 1241 |
+
"step": 17600
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.05765415973133813,
|
| 1245 |
+
"grad_norm": 0.8325287103652954,
|
| 1246 |
+
"learning_rate": 4.7118826813640704e-05,
|
| 1247 |
+
"loss": 0.8886,
|
| 1248 |
+
"step": 17700
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.05797988944733439,
|
| 1252 |
+
"grad_norm": 0.5068339705467224,
|
| 1253 |
+
"learning_rate": 4.710253979732437e-05,
|
| 1254 |
+
"loss": 0.8767,
|
| 1255 |
+
"step": 17800
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.05830561916333065,
|
| 1259 |
+
"grad_norm": 0.7535611391067505,
|
| 1260 |
+
"learning_rate": 4.7086252781008036e-05,
|
| 1261 |
+
"loss": 0.8661,
|
| 1262 |
+
"step": 17900
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.05863134887932691,
|
| 1266 |
+
"grad_norm": 0.9104974865913391,
|
| 1267 |
+
"learning_rate": 4.70699657646917e-05,
|
| 1268 |
+
"loss": 0.8612,
|
| 1269 |
+
"step": 18000
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.058957078595323176,
|
| 1273 |
+
"grad_norm": 0.9106101989746094,
|
| 1274 |
+
"learning_rate": 4.7053678748375374e-05,
|
| 1275 |
+
"loss": 0.8885,
|
| 1276 |
+
"step": 18100
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.059282808311319435,
|
| 1280 |
+
"grad_norm": 0.9990994334220886,
|
| 1281 |
+
"learning_rate": 4.703739173205904e-05,
|
| 1282 |
+
"loss": 0.9097,
|
| 1283 |
+
"step": 18200
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.059608538027315694,
|
| 1287 |
+
"grad_norm": 0.6219133138656616,
|
| 1288 |
+
"learning_rate": 4.7021104715742705e-05,
|
| 1289 |
+
"loss": 0.8349,
|
| 1290 |
+
"step": 18300
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.05993426774331195,
|
| 1294 |
+
"grad_norm": 0.28884798288345337,
|
| 1295 |
+
"learning_rate": 4.700481769942637e-05,
|
| 1296 |
+
"loss": 0.8359,
|
| 1297 |
+
"step": 18400
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.06025999745930821,
|
| 1301 |
+
"grad_norm": 0.6142743229866028,
|
| 1302 |
+
"learning_rate": 4.698853068311004e-05,
|
| 1303 |
+
"loss": 0.8686,
|
| 1304 |
+
"step": 18500
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.06058572717530448,
|
| 1308 |
+
"grad_norm": 0.7121238708496094,
|
| 1309 |
+
"learning_rate": 4.697224366679371e-05,
|
| 1310 |
+
"loss": 0.8318,
|
| 1311 |
+
"step": 18600
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.06091145689130074,
|
| 1315 |
+
"grad_norm": 0.3502013683319092,
|
| 1316 |
+
"learning_rate": 4.6955956650477375e-05,
|
| 1317 |
+
"loss": 0.8353,
|
| 1318 |
+
"step": 18700
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.061237186607296996,
|
| 1322 |
+
"grad_norm": 0.869159460067749,
|
| 1323 |
+
"learning_rate": 4.693966963416104e-05,
|
| 1324 |
+
"loss": 0.8811,
|
| 1325 |
+
"step": 18800
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.061562916323293256,
|
| 1329 |
+
"grad_norm": 0.4008027911186218,
|
| 1330 |
+
"learning_rate": 4.6923382617844706e-05,
|
| 1331 |
+
"loss": 0.8595,
|
| 1332 |
+
"step": 18900
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.06188864603928952,
|
| 1336 |
+
"grad_norm": 0.6609760522842407,
|
| 1337 |
+
"learning_rate": 4.690709560152838e-05,
|
| 1338 |
+
"loss": 0.8591,
|
| 1339 |
+
"step": 19000
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.06221437575528578,
|
| 1343 |
+
"grad_norm": 0.41599878668785095,
|
| 1344 |
+
"learning_rate": 4.6890808585212045e-05,
|
| 1345 |
+
"loss": 0.8792,
|
| 1346 |
+
"step": 19100
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.06254010547128204,
|
| 1350 |
+
"grad_norm": 0.8219528794288635,
|
| 1351 |
+
"learning_rate": 4.687452156889571e-05,
|
| 1352 |
+
"loss": 0.8469,
|
| 1353 |
+
"step": 19200
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.0628658351872783,
|
| 1357 |
+
"grad_norm": 0.5383628010749817,
|
| 1358 |
+
"learning_rate": 4.6858234552579376e-05,
|
| 1359 |
+
"loss": 0.8619,
|
| 1360 |
+
"step": 19300
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.06319156490327456,
|
| 1364 |
+
"grad_norm": 1.0892442464828491,
|
| 1365 |
+
"learning_rate": 4.684194753626304e-05,
|
| 1366 |
+
"loss": 0.8219,
|
| 1367 |
+
"step": 19400
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.06351729461927082,
|
| 1371 |
+
"grad_norm": 0.7258702516555786,
|
| 1372 |
+
"learning_rate": 4.6825660519946714e-05,
|
| 1373 |
+
"loss": 0.8243,
|
| 1374 |
+
"step": 19500
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.06384302433526708,
|
| 1378 |
+
"grad_norm": 1.2622634172439575,
|
| 1379 |
+
"learning_rate": 4.680937350363038e-05,
|
| 1380 |
+
"loss": 0.8619,
|
| 1381 |
+
"step": 19600
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.06416875405126335,
|
| 1385 |
+
"grad_norm": 0.3901592195034027,
|
| 1386 |
+
"learning_rate": 4.6793086487314046e-05,
|
| 1387 |
+
"loss": 0.8315,
|
| 1388 |
+
"step": 19700
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.06449448376725961,
|
| 1392 |
+
"grad_norm": 0.5976518392562866,
|
| 1393 |
+
"learning_rate": 4.677679947099771e-05,
|
| 1394 |
+
"loss": 0.8193,
|
| 1395 |
+
"step": 19800
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.06482021348325587,
|
| 1399 |
+
"grad_norm": 1.0668984651565552,
|
| 1400 |
+
"learning_rate": 4.676051245468138e-05,
|
| 1401 |
+
"loss": 0.8381,
|
| 1402 |
+
"step": 19900
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.06514594319925213,
|
| 1406 |
+
"grad_norm": 0.6844903826713562,
|
| 1407 |
+
"learning_rate": 4.674422543836505e-05,
|
| 1408 |
+
"loss": 0.8202,
|
| 1409 |
+
"step": 20000
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.06547167291524839,
|
| 1413 |
+
"grad_norm": 0.6987929344177246,
|
| 1414 |
+
"learning_rate": 4.672793842204871e-05,
|
| 1415 |
+
"loss": 0.844,
|
| 1416 |
+
"step": 20100
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.06579740263124464,
|
| 1420 |
+
"grad_norm": 1.0227413177490234,
|
| 1421 |
+
"learning_rate": 4.671165140573238e-05,
|
| 1422 |
+
"loss": 0.8093,
|
| 1423 |
+
"step": 20200
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.0661231323472409,
|
| 1427 |
+
"grad_norm": 0.5901645421981812,
|
| 1428 |
+
"learning_rate": 4.669536438941605e-05,
|
| 1429 |
+
"loss": 0.8068,
|
| 1430 |
+
"step": 20300
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.06644886206323716,
|
| 1434 |
+
"grad_norm": 0.7951213717460632,
|
| 1435 |
+
"learning_rate": 4.667907737309972e-05,
|
| 1436 |
+
"loss": 0.8581,
|
| 1437 |
+
"step": 20400
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.06677459177923342,
|
| 1441 |
+
"grad_norm": 0.617341160774231,
|
| 1442 |
+
"learning_rate": 4.666279035678338e-05,
|
| 1443 |
+
"loss": 0.8427,
|
| 1444 |
+
"step": 20500
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.0671003214952297,
|
| 1448 |
+
"grad_norm": 0.694558322429657,
|
| 1449 |
+
"learning_rate": 4.6646503340467044e-05,
|
| 1450 |
+
"loss": 0.8619,
|
| 1451 |
+
"step": 20600
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.06742605121122595,
|
| 1455 |
+
"grad_norm": 0.6441329717636108,
|
| 1456 |
+
"learning_rate": 4.663021632415072e-05,
|
| 1457 |
+
"loss": 0.8866,
|
| 1458 |
+
"step": 20700
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.06775178092722221,
|
| 1462 |
+
"grad_norm": 0.46440285444259644,
|
| 1463 |
+
"learning_rate": 4.661392930783438e-05,
|
| 1464 |
+
"loss": 0.8435,
|
| 1465 |
+
"step": 20800
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.06807751064321847,
|
| 1469 |
+
"grad_norm": 0.42911046743392944,
|
| 1470 |
+
"learning_rate": 4.659764229151805e-05,
|
| 1471 |
+
"loss": 0.8145,
|
| 1472 |
+
"step": 20900
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.06840324035921473,
|
| 1476 |
+
"grad_norm": 0.7508918046951294,
|
| 1477 |
+
"learning_rate": 4.6581355275201714e-05,
|
| 1478 |
+
"loss": 0.8576,
|
| 1479 |
+
"step": 21000
|
| 1480 |
+
}
|
| 1481 |
+
],
|
| 1482 |
+
"logging_steps": 100,
|
| 1483 |
+
"max_steps": 307003,
|
| 1484 |
+
"num_input_tokens_seen": 0,
|
| 1485 |
+
"num_train_epochs": 1,
|
| 1486 |
+
"save_steps": 500,
|
| 1487 |
+
"stateful_callbacks": {
|
| 1488 |
+
"TrainerControl": {
|
| 1489 |
+
"args": {
|
| 1490 |
+
"should_epoch_stop": false,
|
| 1491 |
+
"should_evaluate": false,
|
| 1492 |
+
"should_log": false,
|
| 1493 |
+
"should_save": true,
|
| 1494 |
+
"should_training_stop": false
|
| 1495 |
+
},
|
| 1496 |
+
"attributes": {}
|
| 1497 |
+
}
|
| 1498 |
+
},
|
| 1499 |
+
"total_flos": 3.12051889078272e+17,
|
| 1500 |
+
"train_batch_size": 4,
|
| 1501 |
+
"trial_name": null,
|
| 1502 |
+
"trial_params": null
|
| 1503 |
+
}
|
checkpoint-21000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1be2d5594a22f4bff72c544c78dc0a47a5bb1e2a1e204a42972561dc45351391
|
| 3 |
+
size 5176
|