Training in progress, step 19000, checkpoint
Browse files- checkpoint-19000/config.json +30 -0
- checkpoint-19000/generation_config.json +6 -0
- checkpoint-19000/model.safetensors +3 -0
- checkpoint-19000/optimizer.pt +3 -0
- checkpoint-19000/rng_state.pth +3 -0
- checkpoint-19000/scheduler.pt +3 -0
- checkpoint-19000/special_tokens_map.json +24 -0
- checkpoint-19000/tokenizer.json +96 -0
- checkpoint-19000/tokenizer_config.json +38 -0
- checkpoint-19000/trainer_state.json +1363 -0
- checkpoint-19000/training_args.bin +3 -0
checkpoint-19000/config.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "EleutherAI/pythia-31m",
|
| 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": 256,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 1024,
|
| 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-19000/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-19000/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b253d0acc67dea2c6dcba1827d0f0831c6bcaa7375b8b18c22d76c8a819d1e6e
|
| 3 |
+
size 18981072
|
checkpoint-19000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0b0d0091accdd0e734538517a6d6363a67bd789b84f93a0ea8c2dfa766e8d681
|
| 3 |
+
size 38009210
|
checkpoint-19000/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-19000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:70e0d0724f0096ef5292f025d1bcd990ac46d7a52631f2d351a2db925c61ebb5
|
| 3 |
+
size 1064
|
checkpoint-19000/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-19000/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-19000/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-19000/trainer_state.json
ADDED
|
@@ -0,0 +1,1363 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.7426516572858036,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 19000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.003908692933083177,
|
| 13 |
+
"grad_norm": 0.47716134786605835,
|
| 14 |
+
"learning_rate": 4.98240400406663e-05,
|
| 15 |
+
"loss": 1.3868,
|
| 16 |
+
"step": 100
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.007817385866166354,
|
| 20 |
+
"grad_norm": 0.28900113701820374,
|
| 21 |
+
"learning_rate": 4.9628528974739973e-05,
|
| 22 |
+
"loss": 1.3348,
|
| 23 |
+
"step": 200
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.01172607879924953,
|
| 27 |
+
"grad_norm": 0.37652796506881714,
|
| 28 |
+
"learning_rate": 4.943301790881364e-05,
|
| 29 |
+
"loss": 1.3298,
|
| 30 |
+
"step": 300
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.015634771732332707,
|
| 34 |
+
"grad_norm": 0.24992702901363373,
|
| 35 |
+
"learning_rate": 4.923750684288731e-05,
|
| 36 |
+
"loss": 1.3309,
|
| 37 |
+
"step": 400
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.019543464665415886,
|
| 41 |
+
"grad_norm": 0.3237004280090332,
|
| 42 |
+
"learning_rate": 4.904199577696098e-05,
|
| 43 |
+
"loss": 1.3228,
|
| 44 |
+
"step": 500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.02345215759849906,
|
| 48 |
+
"grad_norm": 0.259011447429657,
|
| 49 |
+
"learning_rate": 4.884648471103465e-05,
|
| 50 |
+
"loss": 1.3166,
|
| 51 |
+
"step": 600
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.02736085053158224,
|
| 55 |
+
"grad_norm": 0.34121841192245483,
|
| 56 |
+
"learning_rate": 4.8650973645108314e-05,
|
| 57 |
+
"loss": 1.3145,
|
| 58 |
+
"step": 700
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.031269543464665414,
|
| 62 |
+
"grad_norm": 0.41398781538009644,
|
| 63 |
+
"learning_rate": 4.845546257918198e-05,
|
| 64 |
+
"loss": 1.3081,
|
| 65 |
+
"step": 800
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.03517823639774859,
|
| 69 |
+
"grad_norm": 0.4289904832839966,
|
| 70 |
+
"learning_rate": 4.825995151325565e-05,
|
| 71 |
+
"loss": 1.3073,
|
| 72 |
+
"step": 900
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.03908692933083177,
|
| 76 |
+
"grad_norm": 0.21226634085178375,
|
| 77 |
+
"learning_rate": 4.8064440447329324e-05,
|
| 78 |
+
"loss": 1.3057,
|
| 79 |
+
"step": 1000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.04299562226391495,
|
| 83 |
+
"grad_norm": 0.2929254174232483,
|
| 84 |
+
"learning_rate": 4.786892938140299e-05,
|
| 85 |
+
"loss": 1.3034,
|
| 86 |
+
"step": 1100
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.04690431519699812,
|
| 90 |
+
"grad_norm": 0.22025109827518463,
|
| 91 |
+
"learning_rate": 4.767341831547666e-05,
|
| 92 |
+
"loss": 1.3048,
|
| 93 |
+
"step": 1200
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.0508130081300813,
|
| 97 |
+
"grad_norm": 0.36455097794532776,
|
| 98 |
+
"learning_rate": 4.747790724955033e-05,
|
| 99 |
+
"loss": 1.3028,
|
| 100 |
+
"step": 1300
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.05472170106316448,
|
| 104 |
+
"grad_norm": 0.24527208507061005,
|
| 105 |
+
"learning_rate": 4.728239618362399e-05,
|
| 106 |
+
"loss": 1.2996,
|
| 107 |
+
"step": 1400
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.05863039399624766,
|
| 111 |
+
"grad_norm": 0.33187615871429443,
|
| 112 |
+
"learning_rate": 4.7086885117697664e-05,
|
| 113 |
+
"loss": 1.3001,
|
| 114 |
+
"step": 1500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.06253908692933083,
|
| 118 |
+
"grad_norm": 0.502117931842804,
|
| 119 |
+
"learning_rate": 4.689137405177133e-05,
|
| 120 |
+
"loss": 1.2954,
|
| 121 |
+
"step": 1600
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.06644777986241401,
|
| 125 |
+
"grad_norm": 0.289112389087677,
|
| 126 |
+
"learning_rate": 4.6695862985845e-05,
|
| 127 |
+
"loss": 1.2969,
|
| 128 |
+
"step": 1700
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.07035647279549719,
|
| 132 |
+
"grad_norm": 0.24350133538246155,
|
| 133 |
+
"learning_rate": 4.6500351919918674e-05,
|
| 134 |
+
"loss": 1.2929,
|
| 135 |
+
"step": 1800
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.07426516572858036,
|
| 139 |
+
"grad_norm": 0.2194579690694809,
|
| 140 |
+
"learning_rate": 4.630484085399234e-05,
|
| 141 |
+
"loss": 1.2913,
|
| 142 |
+
"step": 1900
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.07817385866166354,
|
| 146 |
+
"grad_norm": 0.3578532636165619,
|
| 147 |
+
"learning_rate": 4.6109329788066005e-05,
|
| 148 |
+
"loss": 1.2946,
|
| 149 |
+
"step": 2000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.08208255159474671,
|
| 153 |
+
"grad_norm": 0.22988037765026093,
|
| 154 |
+
"learning_rate": 4.591381872213967e-05,
|
| 155 |
+
"loss": 1.2906,
|
| 156 |
+
"step": 2100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.0859912445278299,
|
| 160 |
+
"grad_norm": 0.44038262963294983,
|
| 161 |
+
"learning_rate": 4.571830765621334e-05,
|
| 162 |
+
"loss": 1.2941,
|
| 163 |
+
"step": 2200
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.08989993746091307,
|
| 167 |
+
"grad_norm": 0.27333030104637146,
|
| 168 |
+
"learning_rate": 4.5522796590287015e-05,
|
| 169 |
+
"loss": 1.2903,
|
| 170 |
+
"step": 2300
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.09380863039399624,
|
| 174 |
+
"grad_norm": 0.23196421563625336,
|
| 175 |
+
"learning_rate": 4.532728552436068e-05,
|
| 176 |
+
"loss": 1.2897,
|
| 177 |
+
"step": 2400
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.09771732332707943,
|
| 181 |
+
"grad_norm": 0.25054216384887695,
|
| 182 |
+
"learning_rate": 4.513177445843435e-05,
|
| 183 |
+
"loss": 1.2891,
|
| 184 |
+
"step": 2500
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.1016260162601626,
|
| 188 |
+
"grad_norm": 0.17129705846309662,
|
| 189 |
+
"learning_rate": 4.493626339250802e-05,
|
| 190 |
+
"loss": 1.2874,
|
| 191 |
+
"step": 2600
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.10553470919324578,
|
| 195 |
+
"grad_norm": 0.5260087251663208,
|
| 196 |
+
"learning_rate": 4.474075232658168e-05,
|
| 197 |
+
"loss": 1.289,
|
| 198 |
+
"step": 2700
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.10944340212632896,
|
| 202 |
+
"grad_norm": 0.2251499593257904,
|
| 203 |
+
"learning_rate": 4.4545241260655355e-05,
|
| 204 |
+
"loss": 1.2882,
|
| 205 |
+
"step": 2800
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.11335209505941213,
|
| 209 |
+
"grad_norm": 0.24193763732910156,
|
| 210 |
+
"learning_rate": 4.434973019472902e-05,
|
| 211 |
+
"loss": 1.2886,
|
| 212 |
+
"step": 2900
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.11726078799249531,
|
| 216 |
+
"grad_norm": 0.2361423224210739,
|
| 217 |
+
"learning_rate": 4.415421912880269e-05,
|
| 218 |
+
"loss": 1.2863,
|
| 219 |
+
"step": 3000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.12116948092557848,
|
| 223 |
+
"grad_norm": 0.22438161075115204,
|
| 224 |
+
"learning_rate": 4.3958708062876365e-05,
|
| 225 |
+
"loss": 1.2861,
|
| 226 |
+
"step": 3100
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.12507817385866166,
|
| 230 |
+
"grad_norm": 0.30306315422058105,
|
| 231 |
+
"learning_rate": 4.376319699695003e-05,
|
| 232 |
+
"loss": 1.2838,
|
| 233 |
+
"step": 3200
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.12898686679174484,
|
| 237 |
+
"grad_norm": 0.2526134252548218,
|
| 238 |
+
"learning_rate": 4.35676859310237e-05,
|
| 239 |
+
"loss": 1.288,
|
| 240 |
+
"step": 3300
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.13289555972482803,
|
| 244 |
+
"grad_norm": 0.27540501952171326,
|
| 245 |
+
"learning_rate": 4.337217486509736e-05,
|
| 246 |
+
"loss": 1.2878,
|
| 247 |
+
"step": 3400
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.13680425265791119,
|
| 251 |
+
"grad_norm": 0.2769360840320587,
|
| 252 |
+
"learning_rate": 4.3176663799171034e-05,
|
| 253 |
+
"loss": 1.2844,
|
| 254 |
+
"step": 3500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.14071294559099437,
|
| 258 |
+
"grad_norm": 0.3868384063243866,
|
| 259 |
+
"learning_rate": 4.2981152733244706e-05,
|
| 260 |
+
"loss": 1.2856,
|
| 261 |
+
"step": 3600
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.14462163852407756,
|
| 265 |
+
"grad_norm": 0.2426549792289734,
|
| 266 |
+
"learning_rate": 4.278564166731837e-05,
|
| 267 |
+
"loss": 1.2843,
|
| 268 |
+
"step": 3700
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.1485303314571607,
|
| 272 |
+
"grad_norm": 0.28486019372940063,
|
| 273 |
+
"learning_rate": 4.2590130601392044e-05,
|
| 274 |
+
"loss": 1.2812,
|
| 275 |
+
"step": 3800
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.1524390243902439,
|
| 279 |
+
"grad_norm": 0.3195390999317169,
|
| 280 |
+
"learning_rate": 4.239461953546571e-05,
|
| 281 |
+
"loss": 1.2845,
|
| 282 |
+
"step": 3900
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.15634771732332708,
|
| 286 |
+
"grad_norm": 0.3183465003967285,
|
| 287 |
+
"learning_rate": 4.219910846953938e-05,
|
| 288 |
+
"loss": 1.281,
|
| 289 |
+
"step": 4000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.16025641025641027,
|
| 293 |
+
"grad_norm": 0.30105575919151306,
|
| 294 |
+
"learning_rate": 4.2003597403613047e-05,
|
| 295 |
+
"loss": 1.2817,
|
| 296 |
+
"step": 4100
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.16416510318949343,
|
| 300 |
+
"grad_norm": 0.3548736274242401,
|
| 301 |
+
"learning_rate": 4.180808633768671e-05,
|
| 302 |
+
"loss": 1.2805,
|
| 303 |
+
"step": 4200
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.1680737961225766,
|
| 307 |
+
"grad_norm": 0.21018876135349274,
|
| 308 |
+
"learning_rate": 4.1612575271760384e-05,
|
| 309 |
+
"loss": 1.28,
|
| 310 |
+
"step": 4300
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.1719824890556598,
|
| 314 |
+
"grad_norm": 0.2551437318325043,
|
| 315 |
+
"learning_rate": 4.141706420583405e-05,
|
| 316 |
+
"loss": 1.279,
|
| 317 |
+
"step": 4400
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.17589118198874296,
|
| 321 |
+
"grad_norm": 0.21301493048667908,
|
| 322 |
+
"learning_rate": 4.122155313990772e-05,
|
| 323 |
+
"loss": 1.2802,
|
| 324 |
+
"step": 4500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.17979987492182614,
|
| 328 |
+
"grad_norm": 0.3330891728401184,
|
| 329 |
+
"learning_rate": 4.1026042073981394e-05,
|
| 330 |
+
"loss": 1.2784,
|
| 331 |
+
"step": 4600
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.18370856785490933,
|
| 335 |
+
"grad_norm": 0.20550762116909027,
|
| 336 |
+
"learning_rate": 4.083053100805506e-05,
|
| 337 |
+
"loss": 1.2802,
|
| 338 |
+
"step": 4700
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.18761726078799248,
|
| 342 |
+
"grad_norm": 0.2603992521762848,
|
| 343 |
+
"learning_rate": 4.0635019942128725e-05,
|
| 344 |
+
"loss": 1.2792,
|
| 345 |
+
"step": 4800
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.19152595372107567,
|
| 349 |
+
"grad_norm": 0.17917224764823914,
|
| 350 |
+
"learning_rate": 4.04395088762024e-05,
|
| 351 |
+
"loss": 1.2803,
|
| 352 |
+
"step": 4900
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.19543464665415886,
|
| 356 |
+
"grad_norm": 0.17226773500442505,
|
| 357 |
+
"learning_rate": 4.024399781027606e-05,
|
| 358 |
+
"loss": 1.2775,
|
| 359 |
+
"step": 5000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.199343339587242,
|
| 363 |
+
"grad_norm": 0.28991296887397766,
|
| 364 |
+
"learning_rate": 4.0048486744349735e-05,
|
| 365 |
+
"loss": 1.2801,
|
| 366 |
+
"step": 5100
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.2032520325203252,
|
| 370 |
+
"grad_norm": 0.2766139805316925,
|
| 371 |
+
"learning_rate": 3.98529756784234e-05,
|
| 372 |
+
"loss": 1.2773,
|
| 373 |
+
"step": 5200
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.20716072545340838,
|
| 377 |
+
"grad_norm": 0.3101244866847992,
|
| 378 |
+
"learning_rate": 3.965746461249707e-05,
|
| 379 |
+
"loss": 1.2786,
|
| 380 |
+
"step": 5300
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.21106941838649157,
|
| 384 |
+
"grad_norm": 0.2537230849266052,
|
| 385 |
+
"learning_rate": 3.946195354657074e-05,
|
| 386 |
+
"loss": 1.2782,
|
| 387 |
+
"step": 5400
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.21497811131957473,
|
| 391 |
+
"grad_norm": 0.388886958360672,
|
| 392 |
+
"learning_rate": 3.92664424806444e-05,
|
| 393 |
+
"loss": 1.2766,
|
| 394 |
+
"step": 5500
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.2188868042526579,
|
| 398 |
+
"grad_norm": 0.3374210596084595,
|
| 399 |
+
"learning_rate": 3.9070931414718075e-05,
|
| 400 |
+
"loss": 1.273,
|
| 401 |
+
"step": 5600
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.2227954971857411,
|
| 405 |
+
"grad_norm": 0.34907710552215576,
|
| 406 |
+
"learning_rate": 3.887542034879174e-05,
|
| 407 |
+
"loss": 1.2772,
|
| 408 |
+
"step": 5700
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.22670419011882426,
|
| 412 |
+
"grad_norm": 0.22610360383987427,
|
| 413 |
+
"learning_rate": 3.867990928286541e-05,
|
| 414 |
+
"loss": 1.2744,
|
| 415 |
+
"step": 5800
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.23061288305190744,
|
| 419 |
+
"grad_norm": 0.27353575825691223,
|
| 420 |
+
"learning_rate": 3.8484398216939085e-05,
|
| 421 |
+
"loss": 1.2755,
|
| 422 |
+
"step": 5900
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.23452157598499063,
|
| 426 |
+
"grad_norm": 0.3231126368045807,
|
| 427 |
+
"learning_rate": 3.828888715101275e-05,
|
| 428 |
+
"loss": 1.2746,
|
| 429 |
+
"step": 6000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.23843026891807378,
|
| 433 |
+
"grad_norm": 0.3560299575328827,
|
| 434 |
+
"learning_rate": 3.8093376085086416e-05,
|
| 435 |
+
"loss": 1.2725,
|
| 436 |
+
"step": 6100
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.24233896185115697,
|
| 440 |
+
"grad_norm": 0.2950970530509949,
|
| 441 |
+
"learning_rate": 3.789786501916008e-05,
|
| 442 |
+
"loss": 1.27,
|
| 443 |
+
"step": 6200
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.24624765478424016,
|
| 447 |
+
"grad_norm": 0.25703492760658264,
|
| 448 |
+
"learning_rate": 3.770235395323375e-05,
|
| 449 |
+
"loss": 1.2705,
|
| 450 |
+
"step": 6300
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.2501563477173233,
|
| 454 |
+
"grad_norm": 0.4161667227745056,
|
| 455 |
+
"learning_rate": 3.7506842887307426e-05,
|
| 456 |
+
"loss": 1.2725,
|
| 457 |
+
"step": 6400
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.2540650406504065,
|
| 461 |
+
"grad_norm": 0.4292469322681427,
|
| 462 |
+
"learning_rate": 3.731133182138109e-05,
|
| 463 |
+
"loss": 1.2736,
|
| 464 |
+
"step": 6500
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.2579737335834897,
|
| 468 |
+
"grad_norm": 0.28723031282424927,
|
| 469 |
+
"learning_rate": 3.711582075545476e-05,
|
| 470 |
+
"loss": 1.2696,
|
| 471 |
+
"step": 6600
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.26188242651657284,
|
| 475 |
+
"grad_norm": 0.3981460928916931,
|
| 476 |
+
"learning_rate": 3.692030968952843e-05,
|
| 477 |
+
"loss": 1.2712,
|
| 478 |
+
"step": 6700
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.26579111944965605,
|
| 482 |
+
"grad_norm": 0.509950578212738,
|
| 483 |
+
"learning_rate": 3.6724798623602094e-05,
|
| 484 |
+
"loss": 1.272,
|
| 485 |
+
"step": 6800
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.2696998123827392,
|
| 489 |
+
"grad_norm": 0.2820459008216858,
|
| 490 |
+
"learning_rate": 3.6529287557675766e-05,
|
| 491 |
+
"loss": 1.2704,
|
| 492 |
+
"step": 6900
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.27360850531582237,
|
| 496 |
+
"grad_norm": 0.2927381098270416,
|
| 497 |
+
"learning_rate": 3.633377649174943e-05,
|
| 498 |
+
"loss": 1.268,
|
| 499 |
+
"step": 7000
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.2775171982489056,
|
| 503 |
+
"grad_norm": 0.2999459505081177,
|
| 504 |
+
"learning_rate": 3.6138265425823104e-05,
|
| 505 |
+
"loss": 1.2673,
|
| 506 |
+
"step": 7100
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.28142589118198874,
|
| 510 |
+
"grad_norm": 0.3287661373615265,
|
| 511 |
+
"learning_rate": 3.5942754359896776e-05,
|
| 512 |
+
"loss": 1.2693,
|
| 513 |
+
"step": 7200
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.2853345841150719,
|
| 517 |
+
"grad_norm": 0.4145341217517853,
|
| 518 |
+
"learning_rate": 3.574724329397044e-05,
|
| 519 |
+
"loss": 1.2655,
|
| 520 |
+
"step": 7300
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.2892432770481551,
|
| 524 |
+
"grad_norm": 0.41897234320640564,
|
| 525 |
+
"learning_rate": 3.555173222804411e-05,
|
| 526 |
+
"loss": 1.2633,
|
| 527 |
+
"step": 7400
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.29315196998123827,
|
| 531 |
+
"grad_norm": 0.6244714856147766,
|
| 532 |
+
"learning_rate": 3.535622116211777e-05,
|
| 533 |
+
"loss": 1.2659,
|
| 534 |
+
"step": 7500
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.2970606629143214,
|
| 538 |
+
"grad_norm": 0.41341808438301086,
|
| 539 |
+
"learning_rate": 3.5160710096191444e-05,
|
| 540 |
+
"loss": 1.2635,
|
| 541 |
+
"step": 7600
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.30096935584740464,
|
| 545 |
+
"grad_norm": 0.39480525255203247,
|
| 546 |
+
"learning_rate": 3.4965199030265117e-05,
|
| 547 |
+
"loss": 1.2633,
|
| 548 |
+
"step": 7700
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.3048780487804878,
|
| 552 |
+
"grad_norm": 0.5332251787185669,
|
| 553 |
+
"learning_rate": 3.476968796433878e-05,
|
| 554 |
+
"loss": 1.262,
|
| 555 |
+
"step": 7800
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.30878674171357096,
|
| 559 |
+
"grad_norm": 0.46669408679008484,
|
| 560 |
+
"learning_rate": 3.4574176898412454e-05,
|
| 561 |
+
"loss": 1.2587,
|
| 562 |
+
"step": 7900
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.31269543464665417,
|
| 566 |
+
"grad_norm": 0.46353575587272644,
|
| 567 |
+
"learning_rate": 3.437866583248612e-05,
|
| 568 |
+
"loss": 1.2611,
|
| 569 |
+
"step": 8000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.3166041275797373,
|
| 573 |
+
"grad_norm": 0.5144259333610535,
|
| 574 |
+
"learning_rate": 3.418315476655979e-05,
|
| 575 |
+
"loss": 1.2603,
|
| 576 |
+
"step": 8100
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.32051282051282054,
|
| 580 |
+
"grad_norm": 0.501035749912262,
|
| 581 |
+
"learning_rate": 3.398764370063346e-05,
|
| 582 |
+
"loss": 1.26,
|
| 583 |
+
"step": 8200
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.3244215134459037,
|
| 587 |
+
"grad_norm": 0.6831699013710022,
|
| 588 |
+
"learning_rate": 3.379213263470712e-05,
|
| 589 |
+
"loss": 1.2598,
|
| 590 |
+
"step": 8300
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.32833020637898686,
|
| 594 |
+
"grad_norm": 0.5810164213180542,
|
| 595 |
+
"learning_rate": 3.3596621568780795e-05,
|
| 596 |
+
"loss": 1.2561,
|
| 597 |
+
"step": 8400
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.33223889931207007,
|
| 601 |
+
"grad_norm": 0.45409688353538513,
|
| 602 |
+
"learning_rate": 3.340111050285447e-05,
|
| 603 |
+
"loss": 1.2552,
|
| 604 |
+
"step": 8500
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.3361475922451532,
|
| 608 |
+
"grad_norm": 0.6193764805793762,
|
| 609 |
+
"learning_rate": 3.320559943692813e-05,
|
| 610 |
+
"loss": 1.2538,
|
| 611 |
+
"step": 8600
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.3400562851782364,
|
| 615 |
+
"grad_norm": 0.4400370717048645,
|
| 616 |
+
"learning_rate": 3.3010088371001805e-05,
|
| 617 |
+
"loss": 1.2548,
|
| 618 |
+
"step": 8700
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.3439649781113196,
|
| 622 |
+
"grad_norm": 0.7784037590026855,
|
| 623 |
+
"learning_rate": 3.281457730507547e-05,
|
| 624 |
+
"loss": 1.2537,
|
| 625 |
+
"step": 8800
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.34787367104440275,
|
| 629 |
+
"grad_norm": 0.7589293718338013,
|
| 630 |
+
"learning_rate": 3.2619066239149135e-05,
|
| 631 |
+
"loss": 1.2519,
|
| 632 |
+
"step": 8900
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.3517823639774859,
|
| 636 |
+
"grad_norm": 0.4971660375595093,
|
| 637 |
+
"learning_rate": 3.242355517322281e-05,
|
| 638 |
+
"loss": 1.25,
|
| 639 |
+
"step": 9000
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.3556910569105691,
|
| 643 |
+
"grad_norm": 0.5253425240516663,
|
| 644 |
+
"learning_rate": 3.222804410729647e-05,
|
| 645 |
+
"loss": 1.2497,
|
| 646 |
+
"step": 9100
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.3595997498436523,
|
| 650 |
+
"grad_norm": 0.7438779473304749,
|
| 651 |
+
"learning_rate": 3.2032533041370145e-05,
|
| 652 |
+
"loss": 1.2458,
|
| 653 |
+
"step": 9200
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.36350844277673544,
|
| 657 |
+
"grad_norm": 0.9425673484802246,
|
| 658 |
+
"learning_rate": 3.183702197544381e-05,
|
| 659 |
+
"loss": 1.2422,
|
| 660 |
+
"step": 9300
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.36741713570981865,
|
| 664 |
+
"grad_norm": 0.5804303288459778,
|
| 665 |
+
"learning_rate": 3.164151090951748e-05,
|
| 666 |
+
"loss": 1.2445,
|
| 667 |
+
"step": 9400
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.3713258286429018,
|
| 671 |
+
"grad_norm": 0.7079547047615051,
|
| 672 |
+
"learning_rate": 3.144599984359115e-05,
|
| 673 |
+
"loss": 1.2402,
|
| 674 |
+
"step": 9500
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.37523452157598497,
|
| 678 |
+
"grad_norm": 1.088934063911438,
|
| 679 |
+
"learning_rate": 3.1250488777664814e-05,
|
| 680 |
+
"loss": 1.2453,
|
| 681 |
+
"step": 9600
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.3791432145090682,
|
| 685 |
+
"grad_norm": 0.48442408442497253,
|
| 686 |
+
"learning_rate": 3.1054977711738486e-05,
|
| 687 |
+
"loss": 1.2469,
|
| 688 |
+
"step": 9700
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.38305190744215134,
|
| 692 |
+
"grad_norm": 0.803571879863739,
|
| 693 |
+
"learning_rate": 3.085946664581215e-05,
|
| 694 |
+
"loss": 1.2448,
|
| 695 |
+
"step": 9800
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.3869606003752345,
|
| 699 |
+
"grad_norm": 0.7402758598327637,
|
| 700 |
+
"learning_rate": 3.0663955579885823e-05,
|
| 701 |
+
"loss": 1.239,
|
| 702 |
+
"step": 9900
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.3908692933083177,
|
| 706 |
+
"grad_norm": 0.874411404132843,
|
| 707 |
+
"learning_rate": 3.0468444513959492e-05,
|
| 708 |
+
"loss": 1.2424,
|
| 709 |
+
"step": 10000
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.39477798624140087,
|
| 713 |
+
"grad_norm": 0.8369914293289185,
|
| 714 |
+
"learning_rate": 3.0272933448033158e-05,
|
| 715 |
+
"loss": 1.2406,
|
| 716 |
+
"step": 10100
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.398686679174484,
|
| 720 |
+
"grad_norm": 0.9173115491867065,
|
| 721 |
+
"learning_rate": 3.007742238210683e-05,
|
| 722 |
+
"loss": 1.2338,
|
| 723 |
+
"step": 10200
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.40259537210756724,
|
| 727 |
+
"grad_norm": 1.1409565210342407,
|
| 728 |
+
"learning_rate": 2.98819113161805e-05,
|
| 729 |
+
"loss": 1.2332,
|
| 730 |
+
"step": 10300
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.4065040650406504,
|
| 734 |
+
"grad_norm": 1.2753782272338867,
|
| 735 |
+
"learning_rate": 2.9686400250254164e-05,
|
| 736 |
+
"loss": 1.23,
|
| 737 |
+
"step": 10400
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.4104127579737336,
|
| 741 |
+
"grad_norm": 1.3038467168807983,
|
| 742 |
+
"learning_rate": 2.9490889184327836e-05,
|
| 743 |
+
"loss": 1.229,
|
| 744 |
+
"step": 10500
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.41432145090681677,
|
| 748 |
+
"grad_norm": 0.7983734607696533,
|
| 749 |
+
"learning_rate": 2.92953781184015e-05,
|
| 750 |
+
"loss": 1.2318,
|
| 751 |
+
"step": 10600
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.4182301438398999,
|
| 755 |
+
"grad_norm": 0.8295246362686157,
|
| 756 |
+
"learning_rate": 2.909986705247517e-05,
|
| 757 |
+
"loss": 1.2335,
|
| 758 |
+
"step": 10700
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.42213883677298314,
|
| 762 |
+
"grad_norm": 0.8072339296340942,
|
| 763 |
+
"learning_rate": 2.8904355986548843e-05,
|
| 764 |
+
"loss": 1.2241,
|
| 765 |
+
"step": 10800
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.4260475297060663,
|
| 769 |
+
"grad_norm": 0.7493770122528076,
|
| 770 |
+
"learning_rate": 2.8708844920622508e-05,
|
| 771 |
+
"loss": 1.2269,
|
| 772 |
+
"step": 10900
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.42995622263914945,
|
| 776 |
+
"grad_norm": 0.705977737903595,
|
| 777 |
+
"learning_rate": 2.8513333854696177e-05,
|
| 778 |
+
"loss": 1.225,
|
| 779 |
+
"step": 11000
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.43386491557223267,
|
| 783 |
+
"grad_norm": 1.4560986757278442,
|
| 784 |
+
"learning_rate": 2.8317822788769842e-05,
|
| 785 |
+
"loss": 1.2255,
|
| 786 |
+
"step": 11100
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.4377736085053158,
|
| 790 |
+
"grad_norm": 1.0365822315216064,
|
| 791 |
+
"learning_rate": 2.8122311722843514e-05,
|
| 792 |
+
"loss": 1.2228,
|
| 793 |
+
"step": 11200
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.441682301438399,
|
| 797 |
+
"grad_norm": 0.9956429600715637,
|
| 798 |
+
"learning_rate": 2.7926800656917183e-05,
|
| 799 |
+
"loss": 1.2185,
|
| 800 |
+
"step": 11300
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.4455909943714822,
|
| 804 |
+
"grad_norm": 0.9062424302101135,
|
| 805 |
+
"learning_rate": 2.773128959099085e-05,
|
| 806 |
+
"loss": 1.2229,
|
| 807 |
+
"step": 11400
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.44949968730456535,
|
| 811 |
+
"grad_norm": 1.64096999168396,
|
| 812 |
+
"learning_rate": 2.753577852506452e-05,
|
| 813 |
+
"loss": 1.2186,
|
| 814 |
+
"step": 11500
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.4534083802376485,
|
| 818 |
+
"grad_norm": 1.88444983959198,
|
| 819 |
+
"learning_rate": 2.7340267459138186e-05,
|
| 820 |
+
"loss": 1.2206,
|
| 821 |
+
"step": 11600
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.4573170731707317,
|
| 825 |
+
"grad_norm": 1.2047277688980103,
|
| 826 |
+
"learning_rate": 2.7144756393211855e-05,
|
| 827 |
+
"loss": 1.2166,
|
| 828 |
+
"step": 11700
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.4612257661038149,
|
| 832 |
+
"grad_norm": 1.3528029918670654,
|
| 833 |
+
"learning_rate": 2.6949245327285527e-05,
|
| 834 |
+
"loss": 1.2052,
|
| 835 |
+
"step": 11800
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.46513445903689804,
|
| 839 |
+
"grad_norm": 0.8786093592643738,
|
| 840 |
+
"learning_rate": 2.6753734261359193e-05,
|
| 841 |
+
"loss": 1.2166,
|
| 842 |
+
"step": 11900
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.46904315196998125,
|
| 846 |
+
"grad_norm": 1.253136157989502,
|
| 847 |
+
"learning_rate": 2.655822319543286e-05,
|
| 848 |
+
"loss": 1.2184,
|
| 849 |
+
"step": 12000
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.4729518449030644,
|
| 853 |
+
"grad_norm": 1.228351354598999,
|
| 854 |
+
"learning_rate": 2.6362712129506534e-05,
|
| 855 |
+
"loss": 1.2095,
|
| 856 |
+
"step": 12100
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.47686053783614757,
|
| 860 |
+
"grad_norm": 1.020992398262024,
|
| 861 |
+
"learning_rate": 2.61672010635802e-05,
|
| 862 |
+
"loss": 1.2028,
|
| 863 |
+
"step": 12200
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.4807692307692308,
|
| 867 |
+
"grad_norm": 1.2239114046096802,
|
| 868 |
+
"learning_rate": 2.597168999765387e-05,
|
| 869 |
+
"loss": 1.2082,
|
| 870 |
+
"step": 12300
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.48467792370231394,
|
| 874 |
+
"grad_norm": 0.8391302824020386,
|
| 875 |
+
"learning_rate": 2.5776178931727533e-05,
|
| 876 |
+
"loss": 1.2115,
|
| 877 |
+
"step": 12400
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.4885866166353971,
|
| 881 |
+
"grad_norm": 1.241330623626709,
|
| 882 |
+
"learning_rate": 2.5580667865801205e-05,
|
| 883 |
+
"loss": 1.2026,
|
| 884 |
+
"step": 12500
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.4924953095684803,
|
| 888 |
+
"grad_norm": 1.6619151830673218,
|
| 889 |
+
"learning_rate": 2.5385156799874878e-05,
|
| 890 |
+
"loss": 1.2068,
|
| 891 |
+
"step": 12600
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.49640400250156347,
|
| 895 |
+
"grad_norm": 1.0257009267807007,
|
| 896 |
+
"learning_rate": 2.5189645733948543e-05,
|
| 897 |
+
"loss": 1.2058,
|
| 898 |
+
"step": 12700
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.5003126954346466,
|
| 902 |
+
"grad_norm": 1.6809911727905273,
|
| 903 |
+
"learning_rate": 2.4994134668022212e-05,
|
| 904 |
+
"loss": 1.2003,
|
| 905 |
+
"step": 12800
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.5042213883677298,
|
| 909 |
+
"grad_norm": 0.6175216436386108,
|
| 910 |
+
"learning_rate": 2.479862360209588e-05,
|
| 911 |
+
"loss": 1.2084,
|
| 912 |
+
"step": 12900
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.508130081300813,
|
| 916 |
+
"grad_norm": 1.0972894430160522,
|
| 917 |
+
"learning_rate": 2.460311253616955e-05,
|
| 918 |
+
"loss": 1.1951,
|
| 919 |
+
"step": 13000
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.5120387742338962,
|
| 923 |
+
"grad_norm": 1.5631442070007324,
|
| 924 |
+
"learning_rate": 2.4407601470243215e-05,
|
| 925 |
+
"loss": 1.2004,
|
| 926 |
+
"step": 13100
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.5159474671669794,
|
| 930 |
+
"grad_norm": 1.4894869327545166,
|
| 931 |
+
"learning_rate": 2.4212090404316887e-05,
|
| 932 |
+
"loss": 1.1905,
|
| 933 |
+
"step": 13200
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.5198561601000625,
|
| 937 |
+
"grad_norm": 1.101841926574707,
|
| 938 |
+
"learning_rate": 2.4016579338390556e-05,
|
| 939 |
+
"loss": 1.1967,
|
| 940 |
+
"step": 13300
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.5237648530331457,
|
| 944 |
+
"grad_norm": 1.0999876260757446,
|
| 945 |
+
"learning_rate": 2.382106827246422e-05,
|
| 946 |
+
"loss": 1.1895,
|
| 947 |
+
"step": 13400
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.5276735459662288,
|
| 951 |
+
"grad_norm": 1.0898387432098389,
|
| 952 |
+
"learning_rate": 2.362555720653789e-05,
|
| 953 |
+
"loss": 1.1902,
|
| 954 |
+
"step": 13500
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.5315822388993121,
|
| 958 |
+
"grad_norm": 1.2711913585662842,
|
| 959 |
+
"learning_rate": 2.343004614061156e-05,
|
| 960 |
+
"loss": 1.1957,
|
| 961 |
+
"step": 13600
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.5354909318323953,
|
| 965 |
+
"grad_norm": 1.1854056119918823,
|
| 966 |
+
"learning_rate": 2.3234535074685228e-05,
|
| 967 |
+
"loss": 1.1905,
|
| 968 |
+
"step": 13700
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.5393996247654784,
|
| 972 |
+
"grad_norm": 1.2872633934020996,
|
| 973 |
+
"learning_rate": 2.3039024008758896e-05,
|
| 974 |
+
"loss": 1.1908,
|
| 975 |
+
"step": 13800
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.5433083176985616,
|
| 979 |
+
"grad_norm": 1.082897663116455,
|
| 980 |
+
"learning_rate": 2.2843512942832565e-05,
|
| 981 |
+
"loss": 1.1822,
|
| 982 |
+
"step": 13900
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.5472170106316447,
|
| 986 |
+
"grad_norm": 1.7656800746917725,
|
| 987 |
+
"learning_rate": 2.2648001876906234e-05,
|
| 988 |
+
"loss": 1.1883,
|
| 989 |
+
"step": 14000
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.551125703564728,
|
| 993 |
+
"grad_norm": 2.594040632247925,
|
| 994 |
+
"learning_rate": 2.2452490810979903e-05,
|
| 995 |
+
"loss": 1.1881,
|
| 996 |
+
"step": 14100
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.5550343964978112,
|
| 1000 |
+
"grad_norm": 1.6944106817245483,
|
| 1001 |
+
"learning_rate": 2.225697974505357e-05,
|
| 1002 |
+
"loss": 1.1869,
|
| 1003 |
+
"step": 14200
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.5589430894308943,
|
| 1007 |
+
"grad_norm": 1.4870433807373047,
|
| 1008 |
+
"learning_rate": 2.206146867912724e-05,
|
| 1009 |
+
"loss": 1.1862,
|
| 1010 |
+
"step": 14300
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.5628517823639775,
|
| 1014 |
+
"grad_norm": 2.2079107761383057,
|
| 1015 |
+
"learning_rate": 2.1865957613200906e-05,
|
| 1016 |
+
"loss": 1.1794,
|
| 1017 |
+
"step": 14400
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.5667604752970606,
|
| 1021 |
+
"grad_norm": 1.419458270072937,
|
| 1022 |
+
"learning_rate": 2.1670446547274575e-05,
|
| 1023 |
+
"loss": 1.1798,
|
| 1024 |
+
"step": 14500
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.5706691682301438,
|
| 1028 |
+
"grad_norm": 1.2365930080413818,
|
| 1029 |
+
"learning_rate": 2.1474935481348247e-05,
|
| 1030 |
+
"loss": 1.1827,
|
| 1031 |
+
"step": 14600
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.5745778611632271,
|
| 1035 |
+
"grad_norm": 1.2718974351882935,
|
| 1036 |
+
"learning_rate": 2.1279424415421916e-05,
|
| 1037 |
+
"loss": 1.1805,
|
| 1038 |
+
"step": 14700
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.5784865540963102,
|
| 1042 |
+
"grad_norm": 1.7296931743621826,
|
| 1043 |
+
"learning_rate": 2.108391334949558e-05,
|
| 1044 |
+
"loss": 1.1797,
|
| 1045 |
+
"step": 14800
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.5823952470293934,
|
| 1049 |
+
"grad_norm": 0.8652032613754272,
|
| 1050 |
+
"learning_rate": 2.088840228356925e-05,
|
| 1051 |
+
"loss": 1.1719,
|
| 1052 |
+
"step": 14900
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.5863039399624765,
|
| 1056 |
+
"grad_norm": 2.346026659011841,
|
| 1057 |
+
"learning_rate": 2.0692891217642922e-05,
|
| 1058 |
+
"loss": 1.1847,
|
| 1059 |
+
"step": 15000
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.5902126328955597,
|
| 1063 |
+
"grad_norm": 1.951084852218628,
|
| 1064 |
+
"learning_rate": 2.0497380151716587e-05,
|
| 1065 |
+
"loss": 1.1794,
|
| 1066 |
+
"step": 15100
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.5941213258286429,
|
| 1070 |
+
"grad_norm": 1.3734288215637207,
|
| 1071 |
+
"learning_rate": 2.0301869085790256e-05,
|
| 1072 |
+
"loss": 1.1749,
|
| 1073 |
+
"step": 15200
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.5980300187617261,
|
| 1077 |
+
"grad_norm": 2.9074900150299072,
|
| 1078 |
+
"learning_rate": 2.0106358019863925e-05,
|
| 1079 |
+
"loss": 1.1729,
|
| 1080 |
+
"step": 15300
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.6019387116948093,
|
| 1084 |
+
"grad_norm": 2.278609275817871,
|
| 1085 |
+
"learning_rate": 1.9910846953937594e-05,
|
| 1086 |
+
"loss": 1.1703,
|
| 1087 |
+
"step": 15400
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.6058474046278924,
|
| 1091 |
+
"grad_norm": 1.291671633720398,
|
| 1092 |
+
"learning_rate": 1.9715335888011263e-05,
|
| 1093 |
+
"loss": 1.1621,
|
| 1094 |
+
"step": 15500
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.6097560975609756,
|
| 1098 |
+
"grad_norm": 1.9885281324386597,
|
| 1099 |
+
"learning_rate": 1.951982482208493e-05,
|
| 1100 |
+
"loss": 1.1636,
|
| 1101 |
+
"step": 15600
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.6136647904940588,
|
| 1105 |
+
"grad_norm": 1.5937877893447876,
|
| 1106 |
+
"learning_rate": 1.93243137561586e-05,
|
| 1107 |
+
"loss": 1.1701,
|
| 1108 |
+
"step": 15700
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.6175734834271419,
|
| 1112 |
+
"grad_norm": 2.537714719772339,
|
| 1113 |
+
"learning_rate": 1.9128802690232266e-05,
|
| 1114 |
+
"loss": 1.1659,
|
| 1115 |
+
"step": 15800
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.6214821763602252,
|
| 1119 |
+
"grad_norm": 1.8786745071411133,
|
| 1120 |
+
"learning_rate": 1.8933291624305938e-05,
|
| 1121 |
+
"loss": 1.163,
|
| 1122 |
+
"step": 15900
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.6253908692933083,
|
| 1126 |
+
"grad_norm": 1.65304696559906,
|
| 1127 |
+
"learning_rate": 1.8737780558379607e-05,
|
| 1128 |
+
"loss": 1.1624,
|
| 1129 |
+
"step": 16000
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.6292995622263915,
|
| 1133 |
+
"grad_norm": 1.7467074394226074,
|
| 1134 |
+
"learning_rate": 1.8542269492453272e-05,
|
| 1135 |
+
"loss": 1.1672,
|
| 1136 |
+
"step": 16100
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.6332082551594747,
|
| 1140 |
+
"grad_norm": 3.616380214691162,
|
| 1141 |
+
"learning_rate": 1.834675842652694e-05,
|
| 1142 |
+
"loss": 1.1672,
|
| 1143 |
+
"step": 16200
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.6371169480925578,
|
| 1147 |
+
"grad_norm": 1.736578106880188,
|
| 1148 |
+
"learning_rate": 1.815124736060061e-05,
|
| 1149 |
+
"loss": 1.1556,
|
| 1150 |
+
"step": 16300
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.6410256410256411,
|
| 1154 |
+
"grad_norm": 1.544382929801941,
|
| 1155 |
+
"learning_rate": 1.7955736294674282e-05,
|
| 1156 |
+
"loss": 1.1601,
|
| 1157 |
+
"step": 16400
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.6449343339587242,
|
| 1161 |
+
"grad_norm": 1.1988792419433594,
|
| 1162 |
+
"learning_rate": 1.7760225228747947e-05,
|
| 1163 |
+
"loss": 1.1568,
|
| 1164 |
+
"step": 16500
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.6488430268918074,
|
| 1168 |
+
"grad_norm": 2.0486536026000977,
|
| 1169 |
+
"learning_rate": 1.7564714162821616e-05,
|
| 1170 |
+
"loss": 1.1522,
|
| 1171 |
+
"step": 16600
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.6527517198248906,
|
| 1175 |
+
"grad_norm": 1.95686674118042,
|
| 1176 |
+
"learning_rate": 1.7369203096895285e-05,
|
| 1177 |
+
"loss": 1.1577,
|
| 1178 |
+
"step": 16700
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.6566604127579737,
|
| 1182 |
+
"grad_norm": 1.7548372745513916,
|
| 1183 |
+
"learning_rate": 1.7173692030968954e-05,
|
| 1184 |
+
"loss": 1.1589,
|
| 1185 |
+
"step": 16800
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.6605691056910569,
|
| 1189 |
+
"grad_norm": 1.1432212591171265,
|
| 1190 |
+
"learning_rate": 1.6978180965042622e-05,
|
| 1191 |
+
"loss": 1.1429,
|
| 1192 |
+
"step": 16900
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.6644777986241401,
|
| 1196 |
+
"grad_norm": 1.4253298044204712,
|
| 1197 |
+
"learning_rate": 1.678266989911629e-05,
|
| 1198 |
+
"loss": 1.1538,
|
| 1199 |
+
"step": 17000
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.6683864915572233,
|
| 1203 |
+
"grad_norm": 2.0788207054138184,
|
| 1204 |
+
"learning_rate": 1.658715883318996e-05,
|
| 1205 |
+
"loss": 1.1441,
|
| 1206 |
+
"step": 17100
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.6722951844903065,
|
| 1210 |
+
"grad_norm": 2.1387529373168945,
|
| 1211 |
+
"learning_rate": 1.639164776726363e-05,
|
| 1212 |
+
"loss": 1.1476,
|
| 1213 |
+
"step": 17200
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.6762038774233896,
|
| 1217 |
+
"grad_norm": 1.2654587030410767,
|
| 1218 |
+
"learning_rate": 1.6196136701337298e-05,
|
| 1219 |
+
"loss": 1.1447,
|
| 1220 |
+
"step": 17300
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.6801125703564728,
|
| 1224 |
+
"grad_norm": 2.0427193641662598,
|
| 1225 |
+
"learning_rate": 1.6000625635410967e-05,
|
| 1226 |
+
"loss": 1.157,
|
| 1227 |
+
"step": 17400
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.6840212632895559,
|
| 1231 |
+
"grad_norm": 1.7374306917190552,
|
| 1232 |
+
"learning_rate": 1.5805114569484632e-05,
|
| 1233 |
+
"loss": 1.1565,
|
| 1234 |
+
"step": 17500
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.6879299562226392,
|
| 1238 |
+
"grad_norm": 2.3408329486846924,
|
| 1239 |
+
"learning_rate": 1.56096035035583e-05,
|
| 1240 |
+
"loss": 1.1446,
|
| 1241 |
+
"step": 17600
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.6918386491557224,
|
| 1245 |
+
"grad_norm": 1.5187078714370728,
|
| 1246 |
+
"learning_rate": 1.5414092437631973e-05,
|
| 1247 |
+
"loss": 1.1558,
|
| 1248 |
+
"step": 17700
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.6957473420888055,
|
| 1252 |
+
"grad_norm": 2.5938124656677246,
|
| 1253 |
+
"learning_rate": 1.521858137170564e-05,
|
| 1254 |
+
"loss": 1.1489,
|
| 1255 |
+
"step": 17800
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.6996560350218887,
|
| 1259 |
+
"grad_norm": 2.978803873062134,
|
| 1260 |
+
"learning_rate": 1.5023070305779307e-05,
|
| 1261 |
+
"loss": 1.1412,
|
| 1262 |
+
"step": 17900
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.7035647279549718,
|
| 1266 |
+
"grad_norm": 3.2251853942871094,
|
| 1267 |
+
"learning_rate": 1.4827559239852976e-05,
|
| 1268 |
+
"loss": 1.1472,
|
| 1269 |
+
"step": 18000
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.707473420888055,
|
| 1273 |
+
"grad_norm": 1.528054118156433,
|
| 1274 |
+
"learning_rate": 1.4632048173926646e-05,
|
| 1275 |
+
"loss": 1.1406,
|
| 1276 |
+
"step": 18100
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.7113821138211383,
|
| 1280 |
+
"grad_norm": 1.627551794052124,
|
| 1281 |
+
"learning_rate": 1.4436537108000314e-05,
|
| 1282 |
+
"loss": 1.1378,
|
| 1283 |
+
"step": 18200
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.7152908067542214,
|
| 1287 |
+
"grad_norm": 1.2601124048233032,
|
| 1288 |
+
"learning_rate": 1.4241026042073982e-05,
|
| 1289 |
+
"loss": 1.1423,
|
| 1290 |
+
"step": 18300
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.7191994996873046,
|
| 1294 |
+
"grad_norm": 1.6382373571395874,
|
| 1295 |
+
"learning_rate": 1.404551497614765e-05,
|
| 1296 |
+
"loss": 1.1397,
|
| 1297 |
+
"step": 18400
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.7231081926203877,
|
| 1301 |
+
"grad_norm": 2.1206157207489014,
|
| 1302 |
+
"learning_rate": 1.3850003910221318e-05,
|
| 1303 |
+
"loss": 1.1401,
|
| 1304 |
+
"step": 18500
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.7270168855534709,
|
| 1308 |
+
"grad_norm": 1.3527886867523193,
|
| 1309 |
+
"learning_rate": 1.3654492844294989e-05,
|
| 1310 |
+
"loss": 1.142,
|
| 1311 |
+
"step": 18600
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.7309255784865542,
|
| 1315 |
+
"grad_norm": 1.8993370532989502,
|
| 1316 |
+
"learning_rate": 1.3458981778368658e-05,
|
| 1317 |
+
"loss": 1.1403,
|
| 1318 |
+
"step": 18700
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.7348342714196373,
|
| 1322 |
+
"grad_norm": 1.5903937816619873,
|
| 1323 |
+
"learning_rate": 1.3263470712442325e-05,
|
| 1324 |
+
"loss": 1.133,
|
| 1325 |
+
"step": 18800
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.7387429643527205,
|
| 1329 |
+
"grad_norm": 1.9694344997406006,
|
| 1330 |
+
"learning_rate": 1.3067959646515993e-05,
|
| 1331 |
+
"loss": 1.1279,
|
| 1332 |
+
"step": 18900
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.7426516572858036,
|
| 1336 |
+
"grad_norm": 2.955965757369995,
|
| 1337 |
+
"learning_rate": 1.2872448580589664e-05,
|
| 1338 |
+
"loss": 1.1221,
|
| 1339 |
+
"step": 19000
|
| 1340 |
+
}
|
| 1341 |
+
],
|
| 1342 |
+
"logging_steps": 100,
|
| 1343 |
+
"max_steps": 25584,
|
| 1344 |
+
"num_input_tokens_seen": 0,
|
| 1345 |
+
"num_train_epochs": 1,
|
| 1346 |
+
"save_steps": 500,
|
| 1347 |
+
"stateful_callbacks": {
|
| 1348 |
+
"TrainerControl": {
|
| 1349 |
+
"args": {
|
| 1350 |
+
"should_epoch_stop": false,
|
| 1351 |
+
"should_evaluate": false,
|
| 1352 |
+
"should_log": false,
|
| 1353 |
+
"should_save": true,
|
| 1354 |
+
"should_training_stop": false
|
| 1355 |
+
},
|
| 1356 |
+
"attributes": {}
|
| 1357 |
+
}
|
| 1358 |
+
},
|
| 1359 |
+
"total_flos": 5.313210089472e+16,
|
| 1360 |
+
"train_batch_size": 48,
|
| 1361 |
+
"trial_name": null,
|
| 1362 |
+
"trial_params": null
|
| 1363 |
+
}
|
checkpoint-19000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:805b2d71d9e6f4f633c9bcb1e0cb62d12ec75a69c6a27204388e915e32239763
|
| 3 |
+
size 5176
|