Commit
•
74f33c6
1
Parent(s):
143f7cd
Upload 11 files
Browse files- config.json +31 -0
- generation_config.json +7 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +24 -0
- tokenizer.json +0 -0
- tokenizer_config.json +81 -0
- trainer_state.json +1433 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "opt-350m",
|
| 3 |
+
"_remove_final_layer_norm": false,
|
| 4 |
+
"activation_dropout": 0.0,
|
| 5 |
+
"activation_function": "relu",
|
| 6 |
+
"architectures": [
|
| 7 |
+
"OPTForCausalLM"
|
| 8 |
+
],
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"bos_token_id": 2,
|
| 11 |
+
"do_layer_norm_before": false,
|
| 12 |
+
"dropout": 0.1,
|
| 13 |
+
"enable_bias": true,
|
| 14 |
+
"eos_token_id": 2,
|
| 15 |
+
"ffn_dim": 4096,
|
| 16 |
+
"hidden_size": 1024,
|
| 17 |
+
"init_std": 0.02,
|
| 18 |
+
"layer_norm_elementwise_affine": true,
|
| 19 |
+
"layerdrop": 0.0,
|
| 20 |
+
"max_position_embeddings": 1024,
|
| 21 |
+
"model_type": "opt",
|
| 22 |
+
"num_attention_heads": 16,
|
| 23 |
+
"num_hidden_layers": 24,
|
| 24 |
+
"pad_token_id": 1,
|
| 25 |
+
"prefix": "</s>",
|
| 26 |
+
"torch_dtype": "float32",
|
| 27 |
+
"transformers_version": "4.42.4",
|
| 28 |
+
"use_cache": true,
|
| 29 |
+
"vocab_size": 46336,
|
| 30 |
+
"word_embed_proj_dim": 512
|
| 31 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 2,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
+
"pad_token_id": 1,
|
| 6 |
+
"transformers_version": "4.42.4"
|
| 7 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ec2b0ccadb04bccada14a3fc0ccd3d5f3814c53bc945670a8982a20c385ac795
|
| 3 |
+
size 1312575624
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:de8914dcd73f007ad7a3ac7da15a66d4ef7eb5e4462c7688e39ffe2f1d227ca5
|
| 3 |
+
size 2625385566
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d304891207a541ed4fa42ff5b57c1ff31488f6e09c0af457af0f4cad3215934b
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f0fbc8b0570d3fe9a0f03c82d4b4cdc650cfc0d5b5cfc547f4c98ab4966663d0
|
| 3 |
+
size 1064
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "</s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": "<unk>",
|
| 17 |
+
"unk_token": {
|
| 18 |
+
"content": "<unk>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": false,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
}
|
| 24 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,81 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": true,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"add_prefix_space": null,
|
| 5 |
+
"added_tokens_decoder": {
|
| 6 |
+
"0": {
|
| 7 |
+
"content": "<unk>",
|
| 8 |
+
"lstrip": false,
|
| 9 |
+
"normalized": false,
|
| 10 |
+
"rstrip": false,
|
| 11 |
+
"single_word": false,
|
| 12 |
+
"special": true
|
| 13 |
+
},
|
| 14 |
+
"1": {
|
| 15 |
+
"content": "<s>",
|
| 16 |
+
"lstrip": false,
|
| 17 |
+
"normalized": false,
|
| 18 |
+
"rstrip": false,
|
| 19 |
+
"single_word": false,
|
| 20 |
+
"special": true
|
| 21 |
+
},
|
| 22 |
+
"2": {
|
| 23 |
+
"content": "</s>",
|
| 24 |
+
"lstrip": false,
|
| 25 |
+
"normalized": false,
|
| 26 |
+
"rstrip": false,
|
| 27 |
+
"single_word": false,
|
| 28 |
+
"special": true
|
| 29 |
+
},
|
| 30 |
+
"46331": {
|
| 31 |
+
"content": "<|sep|>",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": false,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false,
|
| 36 |
+
"special": true
|
| 37 |
+
},
|
| 38 |
+
"46332": {
|
| 39 |
+
"content": "<|endoftext|>",
|
| 40 |
+
"lstrip": false,
|
| 41 |
+
"normalized": false,
|
| 42 |
+
"rstrip": false,
|
| 43 |
+
"single_word": false,
|
| 44 |
+
"special": true
|
| 45 |
+
},
|
| 46 |
+
"46333": {
|
| 47 |
+
"content": "<|acc|>",
|
| 48 |
+
"lstrip": false,
|
| 49 |
+
"normalized": false,
|
| 50 |
+
"rstrip": false,
|
| 51 |
+
"single_word": false,
|
| 52 |
+
"special": true
|
| 53 |
+
},
|
| 54 |
+
"46334": {
|
| 55 |
+
"content": "<|rrn|>",
|
| 56 |
+
"lstrip": false,
|
| 57 |
+
"normalized": false,
|
| 58 |
+
"rstrip": false,
|
| 59 |
+
"single_word": false,
|
| 60 |
+
"special": true
|
| 61 |
+
},
|
| 62 |
+
"46335": {
|
| 63 |
+
"content": "<|tel|>",
|
| 64 |
+
"lstrip": false,
|
| 65 |
+
"normalized": false,
|
| 66 |
+
"rstrip": false,
|
| 67 |
+
"single_word": false,
|
| 68 |
+
"special": true
|
| 69 |
+
}
|
| 70 |
+
},
|
| 71 |
+
"bos_token": "<s>",
|
| 72 |
+
"clean_up_tokenization_spaces": false,
|
| 73 |
+
"eos_token": "</s>",
|
| 74 |
+
"legacy": false,
|
| 75 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 76 |
+
"pad_token": "<unk>",
|
| 77 |
+
"sp_model_kwargs": {},
|
| 78 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 79 |
+
"unk_token": "<unk>",
|
| 80 |
+
"use_default_system_prompt": false
|
| 81 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1433 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.7414272474513438,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 200,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0037071362372567192,
|
| 13 |
+
"grad_norm": 167957.515625,
|
| 14 |
+
"learning_rate": 0.0003992,
|
| 15 |
+
"loss": 10.791,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0074142724745134385,
|
| 20 |
+
"grad_norm": 109901.1953125,
|
| 21 |
+
"learning_rate": 0.00039840000000000003,
|
| 22 |
+
"loss": 10.4195,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.011121408711770158,
|
| 27 |
+
"grad_norm": 101352.265625,
|
| 28 |
+
"learning_rate": 0.0003976,
|
| 29 |
+
"loss": 10.0242,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.014828544949026877,
|
| 34 |
+
"grad_norm": 105144.921875,
|
| 35 |
+
"learning_rate": 0.0003968,
|
| 36 |
+
"loss": 9.5684,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.018535681186283594,
|
| 41 |
+
"grad_norm": 107496.6640625,
|
| 42 |
+
"learning_rate": 0.00039600000000000003,
|
| 43 |
+
"loss": 9.1775,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.022242817423540315,
|
| 48 |
+
"grad_norm": 99089.9453125,
|
| 49 |
+
"learning_rate": 0.0003952,
|
| 50 |
+
"loss": 8.9319,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.025949953660797033,
|
| 55 |
+
"grad_norm": 91537.4765625,
|
| 56 |
+
"learning_rate": 0.0003944,
|
| 57 |
+
"loss": 8.4907,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.029657089898053754,
|
| 62 |
+
"grad_norm": 113517.7265625,
|
| 63 |
+
"learning_rate": 0.0003936,
|
| 64 |
+
"loss": 8.1206,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.033364226135310475,
|
| 69 |
+
"grad_norm": 120697.0546875,
|
| 70 |
+
"learning_rate": 0.0003928,
|
| 71 |
+
"loss": 8.0377,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.03707136237256719,
|
| 76 |
+
"grad_norm": 115909.0546875,
|
| 77 |
+
"learning_rate": 0.000392,
|
| 78 |
+
"loss": 7.9754,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.04077849860982391,
|
| 83 |
+
"grad_norm": 116857.3359375,
|
| 84 |
+
"learning_rate": 0.0003912,
|
| 85 |
+
"loss": 7.8788,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.04448563484708063,
|
| 90 |
+
"grad_norm": 101768.7109375,
|
| 91 |
+
"learning_rate": 0.0003904,
|
| 92 |
+
"loss": 7.8914,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.04819277108433735,
|
| 97 |
+
"grad_norm": 101978.6015625,
|
| 98 |
+
"learning_rate": 0.0003896,
|
| 99 |
+
"loss": 7.8178,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.051899907321594066,
|
| 104 |
+
"grad_norm": 90827.578125,
|
| 105 |
+
"learning_rate": 0.0003888,
|
| 106 |
+
"loss": 7.85,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.05560704355885079,
|
| 111 |
+
"grad_norm": 82672.1640625,
|
| 112 |
+
"learning_rate": 0.000388,
|
| 113 |
+
"loss": 7.8152,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.05931417979610751,
|
| 118 |
+
"grad_norm": 65482.09375,
|
| 119 |
+
"learning_rate": 0.00038720000000000003,
|
| 120 |
+
"loss": 7.8131,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.06302131603336422,
|
| 125 |
+
"grad_norm": 55323.29296875,
|
| 126 |
+
"learning_rate": 0.0003864,
|
| 127 |
+
"loss": 7.6994,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.06672845227062095,
|
| 132 |
+
"grad_norm": 94588.7109375,
|
| 133 |
+
"learning_rate": 0.0003856,
|
| 134 |
+
"loss": 7.8545,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.07043558850787766,
|
| 139 |
+
"grad_norm": 50202.546875,
|
| 140 |
+
"learning_rate": 0.00038480000000000003,
|
| 141 |
+
"loss": 7.75,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.07414272474513438,
|
| 146 |
+
"grad_norm": 48727.80859375,
|
| 147 |
+
"learning_rate": 0.000384,
|
| 148 |
+
"loss": 7.7449,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.0778498609823911,
|
| 153 |
+
"grad_norm": 53795.23046875,
|
| 154 |
+
"learning_rate": 0.0003832,
|
| 155 |
+
"loss": 7.702,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.08155699721964782,
|
| 160 |
+
"grad_norm": 55052.234375,
|
| 161 |
+
"learning_rate": 0.0003824,
|
| 162 |
+
"loss": 7.7048,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.08526413345690455,
|
| 167 |
+
"grad_norm": 35977.5625,
|
| 168 |
+
"learning_rate": 0.0003816,
|
| 169 |
+
"loss": 7.7986,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.08897126969416126,
|
| 174 |
+
"grad_norm": 55099.60546875,
|
| 175 |
+
"learning_rate": 0.0003808,
|
| 176 |
+
"loss": 7.8071,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.09267840593141798,
|
| 181 |
+
"grad_norm": 34977.36328125,
|
| 182 |
+
"learning_rate": 0.00038,
|
| 183 |
+
"loss": 7.8473,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.0963855421686747,
|
| 188 |
+
"grad_norm": 35271.6640625,
|
| 189 |
+
"learning_rate": 0.0003792,
|
| 190 |
+
"loss": 7.7099,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.10009267840593142,
|
| 195 |
+
"grad_norm": 44887.39453125,
|
| 196 |
+
"learning_rate": 0.0003784,
|
| 197 |
+
"loss": 7.617,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.10379981464318813,
|
| 202 |
+
"grad_norm": 41191.33203125,
|
| 203 |
+
"learning_rate": 0.0003776,
|
| 204 |
+
"loss": 7.697,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.10750695088044486,
|
| 209 |
+
"grad_norm": 45202.97265625,
|
| 210 |
+
"learning_rate": 0.0003768,
|
| 211 |
+
"loss": 7.7264,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.11121408711770157,
|
| 216 |
+
"grad_norm": 44944.65234375,
|
| 217 |
+
"learning_rate": 0.000376,
|
| 218 |
+
"loss": 7.7159,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.11492122335495829,
|
| 223 |
+
"grad_norm": 34502.83203125,
|
| 224 |
+
"learning_rate": 0.0003752,
|
| 225 |
+
"loss": 7.7213,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.11862835959221502,
|
| 230 |
+
"grad_norm": 38415.63671875,
|
| 231 |
+
"learning_rate": 0.00037440000000000005,
|
| 232 |
+
"loss": 7.6674,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.12233549582947173,
|
| 237 |
+
"grad_norm": 34140.18359375,
|
| 238 |
+
"learning_rate": 0.00037360000000000003,
|
| 239 |
+
"loss": 7.6627,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.12604263206672844,
|
| 244 |
+
"grad_norm": 27067.009765625,
|
| 245 |
+
"learning_rate": 0.00037280000000000006,
|
| 246 |
+
"loss": 7.6678,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.12974976830398516,
|
| 251 |
+
"grad_norm": 34192.23828125,
|
| 252 |
+
"learning_rate": 0.00037200000000000004,
|
| 253 |
+
"loss": 7.7438,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.1334569045412419,
|
| 258 |
+
"grad_norm": 42940.6953125,
|
| 259 |
+
"learning_rate": 0.0003712,
|
| 260 |
+
"loss": 7.6559,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.1371640407784986,
|
| 265 |
+
"grad_norm": 28908.26171875,
|
| 266 |
+
"learning_rate": 0.00037040000000000006,
|
| 267 |
+
"loss": 7.6763,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.14087117701575533,
|
| 272 |
+
"grad_norm": 46989.23046875,
|
| 273 |
+
"learning_rate": 0.00036960000000000004,
|
| 274 |
+
"loss": 7.6483,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.14457831325301204,
|
| 279 |
+
"grad_norm": 38628.00390625,
|
| 280 |
+
"learning_rate": 0.0003688,
|
| 281 |
+
"loss": 7.5813,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.14828544949026876,
|
| 286 |
+
"grad_norm": 26901.994140625,
|
| 287 |
+
"learning_rate": 0.00036800000000000005,
|
| 288 |
+
"loss": 7.7328,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.1519925857275255,
|
| 293 |
+
"grad_norm": 55413.51953125,
|
| 294 |
+
"learning_rate": 0.00036720000000000004,
|
| 295 |
+
"loss": 7.5977,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.1556997219647822,
|
| 300 |
+
"grad_norm": 38922.68359375,
|
| 301 |
+
"learning_rate": 0.0003664,
|
| 302 |
+
"loss": 7.6575,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.15940685820203893,
|
| 307 |
+
"grad_norm": 49835.87109375,
|
| 308 |
+
"learning_rate": 0.00036560000000000005,
|
| 309 |
+
"loss": 7.7382,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.16311399443929564,
|
| 314 |
+
"grad_norm": 41342.8515625,
|
| 315 |
+
"learning_rate": 0.00036480000000000003,
|
| 316 |
+
"loss": 7.7068,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.16682113067655235,
|
| 321 |
+
"grad_norm": 38896.15625,
|
| 322 |
+
"learning_rate": 0.000364,
|
| 323 |
+
"loss": 7.6614,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.1705282669138091,
|
| 328 |
+
"grad_norm": 29027.955078125,
|
| 329 |
+
"learning_rate": 0.00036320000000000005,
|
| 330 |
+
"loss": 7.728,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.1742354031510658,
|
| 335 |
+
"grad_norm": 33758.0859375,
|
| 336 |
+
"learning_rate": 0.0003624,
|
| 337 |
+
"loss": 7.7392,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.17794253938832252,
|
| 342 |
+
"grad_norm": 29002.869140625,
|
| 343 |
+
"learning_rate": 0.0003616,
|
| 344 |
+
"loss": 7.666,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.18164967562557924,
|
| 349 |
+
"grad_norm": 33393.12890625,
|
| 350 |
+
"learning_rate": 0.00036080000000000004,
|
| 351 |
+
"loss": 7.6067,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.18535681186283595,
|
| 356 |
+
"grad_norm": 39193.51171875,
|
| 357 |
+
"learning_rate": 0.00036,
|
| 358 |
+
"loss": 7.7868,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.18906394810009267,
|
| 363 |
+
"grad_norm": 25982.78125,
|
| 364 |
+
"learning_rate": 0.0003592,
|
| 365 |
+
"loss": 7.7189,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.1927710843373494,
|
| 370 |
+
"grad_norm": 28694.505859375,
|
| 371 |
+
"learning_rate": 0.00035840000000000004,
|
| 372 |
+
"loss": 7.6999,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.19647822057460612,
|
| 377 |
+
"grad_norm": 26356.8828125,
|
| 378 |
+
"learning_rate": 0.0003576,
|
| 379 |
+
"loss": 7.712,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.20018535681186284,
|
| 384 |
+
"grad_norm": 25880.298828125,
|
| 385 |
+
"learning_rate": 0.0003568,
|
| 386 |
+
"loss": 7.7015,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.20389249304911955,
|
| 391 |
+
"grad_norm": 23557.111328125,
|
| 392 |
+
"learning_rate": 0.00035600000000000003,
|
| 393 |
+
"loss": 7.6849,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.20759962928637626,
|
| 398 |
+
"grad_norm": 31365.33203125,
|
| 399 |
+
"learning_rate": 0.0003552,
|
| 400 |
+
"loss": 7.7333,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.211306765523633,
|
| 405 |
+
"grad_norm": 31506.552734375,
|
| 406 |
+
"learning_rate": 0.0003544,
|
| 407 |
+
"loss": 7.7317,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.21501390176088972,
|
| 412 |
+
"grad_norm": 22261.244140625,
|
| 413 |
+
"learning_rate": 0.00035360000000000003,
|
| 414 |
+
"loss": 7.6978,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.21872103799814643,
|
| 419 |
+
"grad_norm": 36267.4921875,
|
| 420 |
+
"learning_rate": 0.0003528,
|
| 421 |
+
"loss": 7.7125,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.22242817423540315,
|
| 426 |
+
"grad_norm": 29624.087890625,
|
| 427 |
+
"learning_rate": 0.00035200000000000005,
|
| 428 |
+
"loss": 7.734,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.22613531047265986,
|
| 433 |
+
"grad_norm": 25301.228515625,
|
| 434 |
+
"learning_rate": 0.0003512,
|
| 435 |
+
"loss": 7.7287,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.22984244670991658,
|
| 440 |
+
"grad_norm": 26147.228515625,
|
| 441 |
+
"learning_rate": 0.0003504,
|
| 442 |
+
"loss": 7.7059,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.23354958294717332,
|
| 447 |
+
"grad_norm": 27329.443359375,
|
| 448 |
+
"learning_rate": 0.00034960000000000004,
|
| 449 |
+
"loss": 7.6798,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.23725671918443003,
|
| 454 |
+
"grad_norm": 23415.9609375,
|
| 455 |
+
"learning_rate": 0.0003488,
|
| 456 |
+
"loss": 7.6968,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.24096385542168675,
|
| 461 |
+
"grad_norm": 23625.1171875,
|
| 462 |
+
"learning_rate": 0.000348,
|
| 463 |
+
"loss": 7.7119,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.24467099165894346,
|
| 468 |
+
"grad_norm": 23805.42578125,
|
| 469 |
+
"learning_rate": 0.00034720000000000004,
|
| 470 |
+
"loss": 7.6473,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.24837812789620017,
|
| 475 |
+
"grad_norm": 47364.8203125,
|
| 476 |
+
"learning_rate": 0.0003464,
|
| 477 |
+
"loss": 7.7921,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.2520852641334569,
|
| 482 |
+
"grad_norm": 29178.279296875,
|
| 483 |
+
"learning_rate": 0.0003456,
|
| 484 |
+
"loss": 7.6958,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.2557924003707136,
|
| 489 |
+
"grad_norm": 26202.958984375,
|
| 490 |
+
"learning_rate": 0.00034480000000000003,
|
| 491 |
+
"loss": 7.7765,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.2594995366079703,
|
| 496 |
+
"grad_norm": 48753.58203125,
|
| 497 |
+
"learning_rate": 0.000344,
|
| 498 |
+
"loss": 7.6496,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.2632066728452271,
|
| 503 |
+
"grad_norm": 24508.509765625,
|
| 504 |
+
"learning_rate": 0.0003432,
|
| 505 |
+
"loss": 7.7125,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.2669138090824838,
|
| 510 |
+
"grad_norm": 33996.55078125,
|
| 511 |
+
"learning_rate": 0.00034240000000000003,
|
| 512 |
+
"loss": 7.6635,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.2706209453197405,
|
| 517 |
+
"grad_norm": 32989.36328125,
|
| 518 |
+
"learning_rate": 0.0003416,
|
| 519 |
+
"loss": 7.6893,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.2743280815569972,
|
| 524 |
+
"grad_norm": 32296.1796875,
|
| 525 |
+
"learning_rate": 0.0003408,
|
| 526 |
+
"loss": 7.6696,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.27803521779425394,
|
| 531 |
+
"grad_norm": 35698.16015625,
|
| 532 |
+
"learning_rate": 0.00034,
|
| 533 |
+
"loss": 7.6713,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.28174235403151066,
|
| 538 |
+
"grad_norm": 25034.283203125,
|
| 539 |
+
"learning_rate": 0.0003392,
|
| 540 |
+
"loss": 7.6629,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.28544949026876737,
|
| 545 |
+
"grad_norm": 36568.65625,
|
| 546 |
+
"learning_rate": 0.0003384,
|
| 547 |
+
"loss": 7.7075,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.2891566265060241,
|
| 552 |
+
"grad_norm": 25048.875,
|
| 553 |
+
"learning_rate": 0.0003376,
|
| 554 |
+
"loss": 7.6727,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.2928637627432808,
|
| 559 |
+
"grad_norm": 25438.61328125,
|
| 560 |
+
"learning_rate": 0.0003368,
|
| 561 |
+
"loss": 7.7028,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.2965708989805375,
|
| 566 |
+
"grad_norm": 27428.9453125,
|
| 567 |
+
"learning_rate": 0.000336,
|
| 568 |
+
"loss": 7.6516,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.3002780352177943,
|
| 573 |
+
"grad_norm": 32185.8125,
|
| 574 |
+
"learning_rate": 0.0003352,
|
| 575 |
+
"loss": 7.7127,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.303985171455051,
|
| 580 |
+
"grad_norm": 28342.439453125,
|
| 581 |
+
"learning_rate": 0.0003344,
|
| 582 |
+
"loss": 7.6461,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.3076923076923077,
|
| 587 |
+
"grad_norm": 22977.4140625,
|
| 588 |
+
"learning_rate": 0.0003336,
|
| 589 |
+
"loss": 7.6348,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.3113994439295644,
|
| 594 |
+
"grad_norm": 28778.767578125,
|
| 595 |
+
"learning_rate": 0.0003328,
|
| 596 |
+
"loss": 7.6299,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.31510658016682114,
|
| 601 |
+
"grad_norm": 21658.966796875,
|
| 602 |
+
"learning_rate": 0.000332,
|
| 603 |
+
"loss": 7.633,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.31881371640407785,
|
| 608 |
+
"grad_norm": 22994.66796875,
|
| 609 |
+
"learning_rate": 0.0003312,
|
| 610 |
+
"loss": 7.648,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.32252085264133457,
|
| 615 |
+
"grad_norm": 23064.05078125,
|
| 616 |
+
"learning_rate": 0.0003304,
|
| 617 |
+
"loss": 7.712,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.3262279888785913,
|
| 622 |
+
"grad_norm": 34689.19140625,
|
| 623 |
+
"learning_rate": 0.0003296,
|
| 624 |
+
"loss": 7.6168,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.329935125115848,
|
| 629 |
+
"grad_norm": 26677.1328125,
|
| 630 |
+
"learning_rate": 0.0003288,
|
| 631 |
+
"loss": 7.6226,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.3336422613531047,
|
| 636 |
+
"grad_norm": 39699.62109375,
|
| 637 |
+
"learning_rate": 0.000328,
|
| 638 |
+
"loss": 7.6465,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.3373493975903614,
|
| 643 |
+
"grad_norm": 47106.6640625,
|
| 644 |
+
"learning_rate": 0.0003272,
|
| 645 |
+
"loss": 7.6884,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.3410565338276182,
|
| 650 |
+
"grad_norm": 30162.638671875,
|
| 651 |
+
"learning_rate": 0.0003264,
|
| 652 |
+
"loss": 7.7695,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.3447636700648749,
|
| 657 |
+
"grad_norm": 40879.01953125,
|
| 658 |
+
"learning_rate": 0.0003256,
|
| 659 |
+
"loss": 7.7253,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.3484708063021316,
|
| 664 |
+
"grad_norm": 56518.4921875,
|
| 665 |
+
"learning_rate": 0.00032480000000000003,
|
| 666 |
+
"loss": 7.6734,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.35217794253938833,
|
| 671 |
+
"grad_norm": 37450.08203125,
|
| 672 |
+
"learning_rate": 0.000324,
|
| 673 |
+
"loss": 7.6897,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.35588507877664505,
|
| 678 |
+
"grad_norm": 28603.978515625,
|
| 679 |
+
"learning_rate": 0.00032320000000000005,
|
| 680 |
+
"loss": 7.7346,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.35959221501390176,
|
| 685 |
+
"grad_norm": 45344.12109375,
|
| 686 |
+
"learning_rate": 0.00032240000000000003,
|
| 687 |
+
"loss": 7.7564,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.3632993512511585,
|
| 692 |
+
"grad_norm": 20206.189453125,
|
| 693 |
+
"learning_rate": 0.0003216,
|
| 694 |
+
"loss": 7.6465,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.3670064874884152,
|
| 699 |
+
"grad_norm": 29952.62890625,
|
| 700 |
+
"learning_rate": 0.00032080000000000005,
|
| 701 |
+
"loss": 7.6581,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.3707136237256719,
|
| 706 |
+
"grad_norm": 24017.02734375,
|
| 707 |
+
"learning_rate": 0.00032,
|
| 708 |
+
"loss": 7.7068,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.3744207599629286,
|
| 713 |
+
"grad_norm": 21995.66796875,
|
| 714 |
+
"learning_rate": 0.0003192,
|
| 715 |
+
"loss": 7.7306,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.37812789620018533,
|
| 720 |
+
"grad_norm": 22698.15625,
|
| 721 |
+
"learning_rate": 0.00031840000000000004,
|
| 722 |
+
"loss": 7.6167,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.3818350324374421,
|
| 727 |
+
"grad_norm": 19390.587890625,
|
| 728 |
+
"learning_rate": 0.0003176,
|
| 729 |
+
"loss": 7.6298,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.3855421686746988,
|
| 734 |
+
"grad_norm": 23548.39453125,
|
| 735 |
+
"learning_rate": 0.00031680000000000006,
|
| 736 |
+
"loss": 7.7148,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.38924930491195553,
|
| 741 |
+
"grad_norm": 25070.564453125,
|
| 742 |
+
"learning_rate": 0.00031600000000000004,
|
| 743 |
+
"loss": 7.8045,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.39295644114921224,
|
| 748 |
+
"grad_norm": 39852.94921875,
|
| 749 |
+
"learning_rate": 0.0003152,
|
| 750 |
+
"loss": 7.6813,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.39666357738646896,
|
| 755 |
+
"grad_norm": 30994.017578125,
|
| 756 |
+
"learning_rate": 0.00031440000000000005,
|
| 757 |
+
"loss": 7.6801,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.40037071362372567,
|
| 762 |
+
"grad_norm": 35010.94140625,
|
| 763 |
+
"learning_rate": 0.00031360000000000003,
|
| 764 |
+
"loss": 7.7625,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.4040778498609824,
|
| 769 |
+
"grad_norm": 32364.001953125,
|
| 770 |
+
"learning_rate": 0.0003128,
|
| 771 |
+
"loss": 7.682,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.4077849860982391,
|
| 776 |
+
"grad_norm": 24475.48828125,
|
| 777 |
+
"learning_rate": 0.00031200000000000005,
|
| 778 |
+
"loss": 7.6953,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.4114921223354958,
|
| 783 |
+
"grad_norm": 28467.2890625,
|
| 784 |
+
"learning_rate": 0.00031120000000000003,
|
| 785 |
+
"loss": 7.7112,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.4151992585727525,
|
| 790 |
+
"grad_norm": 46241.89453125,
|
| 791 |
+
"learning_rate": 0.0003104,
|
| 792 |
+
"loss": 7.625,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.41890639481000924,
|
| 797 |
+
"grad_norm": 25736.814453125,
|
| 798 |
+
"learning_rate": 0.00030960000000000004,
|
| 799 |
+
"loss": 7.6842,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.422613531047266,
|
| 804 |
+
"grad_norm": 25479.744140625,
|
| 805 |
+
"learning_rate": 0.0003088,
|
| 806 |
+
"loss": 7.7131,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.4263206672845227,
|
| 811 |
+
"grad_norm": 32374.447265625,
|
| 812 |
+
"learning_rate": 0.000308,
|
| 813 |
+
"loss": 7.7209,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.43002780352177944,
|
| 818 |
+
"grad_norm": 21930.126953125,
|
| 819 |
+
"learning_rate": 0.00030720000000000004,
|
| 820 |
+
"loss": 7.6593,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.43373493975903615,
|
| 825 |
+
"grad_norm": 22632.013671875,
|
| 826 |
+
"learning_rate": 0.0003064,
|
| 827 |
+
"loss": 7.7121,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.43744207599629287,
|
| 832 |
+
"grad_norm": 21551.6328125,
|
| 833 |
+
"learning_rate": 0.0003056,
|
| 834 |
+
"loss": 7.6504,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.4411492122335496,
|
| 839 |
+
"grad_norm": 24234.326171875,
|
| 840 |
+
"learning_rate": 0.00030480000000000004,
|
| 841 |
+
"loss": 7.7,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.4448563484708063,
|
| 846 |
+
"grad_norm": 27236.205078125,
|
| 847 |
+
"learning_rate": 0.000304,
|
| 848 |
+
"loss": 7.7073,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.448563484708063,
|
| 853 |
+
"grad_norm": 20109.84765625,
|
| 854 |
+
"learning_rate": 0.0003032,
|
| 855 |
+
"loss": 7.642,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.4522706209453197,
|
| 860 |
+
"grad_norm": 20982.546875,
|
| 861 |
+
"learning_rate": 0.00030240000000000003,
|
| 862 |
+
"loss": 7.7092,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.45597775718257644,
|
| 867 |
+
"grad_norm": 30563.40625,
|
| 868 |
+
"learning_rate": 0.0003016,
|
| 869 |
+
"loss": 7.6086,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.45968489341983315,
|
| 874 |
+
"grad_norm": 26537.8828125,
|
| 875 |
+
"learning_rate": 0.0003008,
|
| 876 |
+
"loss": 7.711,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.4633920296570899,
|
| 881 |
+
"grad_norm": 26180.9765625,
|
| 882 |
+
"learning_rate": 0.00030000000000000003,
|
| 883 |
+
"loss": 7.6946,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.46709916589434664,
|
| 888 |
+
"grad_norm": 25894.8828125,
|
| 889 |
+
"learning_rate": 0.0002992,
|
| 890 |
+
"loss": 7.6252,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.47080630213160335,
|
| 895 |
+
"grad_norm": 17775.234375,
|
| 896 |
+
"learning_rate": 0.0002984,
|
| 897 |
+
"loss": 7.7064,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.47451343836886006,
|
| 902 |
+
"grad_norm": 23387.5625,
|
| 903 |
+
"learning_rate": 0.0002976,
|
| 904 |
+
"loss": 7.606,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.4782205746061168,
|
| 909 |
+
"grad_norm": 26294.63671875,
|
| 910 |
+
"learning_rate": 0.0002968,
|
| 911 |
+
"loss": 7.6753,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.4819277108433735,
|
| 916 |
+
"grad_norm": 22350.404296875,
|
| 917 |
+
"learning_rate": 0.000296,
|
| 918 |
+
"loss": 7.6926,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.4856348470806302,
|
| 923 |
+
"grad_norm": 23048.61328125,
|
| 924 |
+
"learning_rate": 0.0002952,
|
| 925 |
+
"loss": 7.6476,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.4893419833178869,
|
| 930 |
+
"grad_norm": 26630.447265625,
|
| 931 |
+
"learning_rate": 0.0002944,
|
| 932 |
+
"loss": 7.7831,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.49304911955514363,
|
| 937 |
+
"grad_norm": 34660.65234375,
|
| 938 |
+
"learning_rate": 0.00029360000000000003,
|
| 939 |
+
"loss": 7.5954,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.49675625579240035,
|
| 944 |
+
"grad_norm": 19611.568359375,
|
| 945 |
+
"learning_rate": 0.0002928,
|
| 946 |
+
"loss": 7.6305,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.5004633920296571,
|
| 951 |
+
"grad_norm": 38032.05078125,
|
| 952 |
+
"learning_rate": 0.000292,
|
| 953 |
+
"loss": 7.725,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.5041705282669138,
|
| 958 |
+
"grad_norm": 26124.802734375,
|
| 959 |
+
"learning_rate": 0.00029120000000000003,
|
| 960 |
+
"loss": 7.6547,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.5078776645041705,
|
| 965 |
+
"grad_norm": 22567.94921875,
|
| 966 |
+
"learning_rate": 0.0002904,
|
| 967 |
+
"loss": 7.7534,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.5115848007414272,
|
| 972 |
+
"grad_norm": 37485.49609375,
|
| 973 |
+
"learning_rate": 0.0002896,
|
| 974 |
+
"loss": 7.6795,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.5152919369786839,
|
| 979 |
+
"grad_norm": 32182.43359375,
|
| 980 |
+
"learning_rate": 0.0002888,
|
| 981 |
+
"loss": 7.7417,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.5189990732159406,
|
| 986 |
+
"grad_norm": 24093.3125,
|
| 987 |
+
"learning_rate": 0.000288,
|
| 988 |
+
"loss": 7.6875,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.5227062094531975,
|
| 993 |
+
"grad_norm": 23480.59765625,
|
| 994 |
+
"learning_rate": 0.0002872,
|
| 995 |
+
"loss": 7.6571,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.5264133456904542,
|
| 1000 |
+
"grad_norm": 34477.796875,
|
| 1001 |
+
"learning_rate": 0.0002864,
|
| 1002 |
+
"loss": 7.6389,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.5301204819277109,
|
| 1007 |
+
"grad_norm": 32023.896484375,
|
| 1008 |
+
"learning_rate": 0.0002856,
|
| 1009 |
+
"loss": 7.7501,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.5338276181649676,
|
| 1014 |
+
"grad_norm": 21589.513671875,
|
| 1015 |
+
"learning_rate": 0.0002848,
|
| 1016 |
+
"loss": 7.6895,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.5375347544022243,
|
| 1021 |
+
"grad_norm": 31786.94921875,
|
| 1022 |
+
"learning_rate": 0.000284,
|
| 1023 |
+
"loss": 7.7106,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.541241890639481,
|
| 1028 |
+
"grad_norm": 31673.8359375,
|
| 1029 |
+
"learning_rate": 0.0002832,
|
| 1030 |
+
"loss": 7.6815,
|
| 1031 |
+
"step": 146
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.5449490268767377,
|
| 1035 |
+
"grad_norm": 17670.734375,
|
| 1036 |
+
"learning_rate": 0.0002824,
|
| 1037 |
+
"loss": 7.6869,
|
| 1038 |
+
"step": 147
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.5486561631139945,
|
| 1042 |
+
"grad_norm": 34063.0703125,
|
| 1043 |
+
"learning_rate": 0.0002816,
|
| 1044 |
+
"loss": 7.7108,
|
| 1045 |
+
"step": 148
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.5523632993512512,
|
| 1049 |
+
"grad_norm": 36702.2734375,
|
| 1050 |
+
"learning_rate": 0.0002808,
|
| 1051 |
+
"loss": 7.7124,
|
| 1052 |
+
"step": 149
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.5560704355885079,
|
| 1056 |
+
"grad_norm": 22709.572265625,
|
| 1057 |
+
"learning_rate": 0.00028,
|
| 1058 |
+
"loss": 7.7326,
|
| 1059 |
+
"step": 150
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.5597775718257646,
|
| 1063 |
+
"grad_norm": 36804.21484375,
|
| 1064 |
+
"learning_rate": 0.0002792,
|
| 1065 |
+
"loss": 7.6414,
|
| 1066 |
+
"step": 151
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.5634847080630213,
|
| 1070 |
+
"grad_norm": 30339.912109375,
|
| 1071 |
+
"learning_rate": 0.0002784,
|
| 1072 |
+
"loss": 7.7337,
|
| 1073 |
+
"step": 152
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.567191844300278,
|
| 1077 |
+
"grad_norm": 31866.80859375,
|
| 1078 |
+
"learning_rate": 0.00027759999999999997,
|
| 1079 |
+
"loss": 7.6208,
|
| 1080 |
+
"step": 153
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.5708989805375347,
|
| 1084 |
+
"grad_norm": 23864.302734375,
|
| 1085 |
+
"learning_rate": 0.0002768,
|
| 1086 |
+
"loss": 7.7083,
|
| 1087 |
+
"step": 154
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.5746061167747915,
|
| 1091 |
+
"grad_norm": 29230.330078125,
|
| 1092 |
+
"learning_rate": 0.000276,
|
| 1093 |
+
"loss": 7.6914,
|
| 1094 |
+
"step": 155
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.5783132530120482,
|
| 1098 |
+
"grad_norm": 21988.8046875,
|
| 1099 |
+
"learning_rate": 0.00027519999999999997,
|
| 1100 |
+
"loss": 7.7157,
|
| 1101 |
+
"step": 156
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.5820203892493049,
|
| 1105 |
+
"grad_norm": 21070.361328125,
|
| 1106 |
+
"learning_rate": 0.00027440000000000006,
|
| 1107 |
+
"loss": 7.6987,
|
| 1108 |
+
"step": 157
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.5857275254865616,
|
| 1112 |
+
"grad_norm": 39177.30859375,
|
| 1113 |
+
"learning_rate": 0.00027360000000000004,
|
| 1114 |
+
"loss": 7.5922,
|
| 1115 |
+
"step": 158
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.5894346617238183,
|
| 1119 |
+
"grad_norm": 20961.755859375,
|
| 1120 |
+
"learning_rate": 0.0002728,
|
| 1121 |
+
"loss": 7.7621,
|
| 1122 |
+
"step": 159
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.593141797961075,
|
| 1126 |
+
"grad_norm": 24547.12890625,
|
| 1127 |
+
"learning_rate": 0.00027200000000000005,
|
| 1128 |
+
"loss": 7.7387,
|
| 1129 |
+
"step": 160
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.5968489341983317,
|
| 1133 |
+
"grad_norm": 17789.8125,
|
| 1134 |
+
"learning_rate": 0.00027120000000000003,
|
| 1135 |
+
"loss": 7.6818,
|
| 1136 |
+
"step": 161
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.6005560704355886,
|
| 1140 |
+
"grad_norm": 21633.90625,
|
| 1141 |
+
"learning_rate": 0.0002704,
|
| 1142 |
+
"loss": 7.6545,
|
| 1143 |
+
"step": 162
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.6042632066728453,
|
| 1147 |
+
"grad_norm": 17543.3046875,
|
| 1148 |
+
"learning_rate": 0.00026960000000000005,
|
| 1149 |
+
"loss": 7.6662,
|
| 1150 |
+
"step": 163
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.607970342910102,
|
| 1154 |
+
"grad_norm": 18747.458984375,
|
| 1155 |
+
"learning_rate": 0.00026880000000000003,
|
| 1156 |
+
"loss": 7.6227,
|
| 1157 |
+
"step": 164
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.6116774791473587,
|
| 1161 |
+
"grad_norm": 22172.224609375,
|
| 1162 |
+
"learning_rate": 0.000268,
|
| 1163 |
+
"loss": 7.6899,
|
| 1164 |
+
"step": 165
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.6153846153846154,
|
| 1168 |
+
"grad_norm": 19154.330078125,
|
| 1169 |
+
"learning_rate": 0.00026720000000000004,
|
| 1170 |
+
"loss": 7.6195,
|
| 1171 |
+
"step": 166
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.6190917516218721,
|
| 1175 |
+
"grad_norm": 20868.43359375,
|
| 1176 |
+
"learning_rate": 0.0002664,
|
| 1177 |
+
"loss": 7.677,
|
| 1178 |
+
"step": 167
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.6227988878591288,
|
| 1182 |
+
"grad_norm": 18564.533203125,
|
| 1183 |
+
"learning_rate": 0.0002656,
|
| 1184 |
+
"loss": 7.696,
|
| 1185 |
+
"step": 168
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.6265060240963856,
|
| 1189 |
+
"grad_norm": 22970.892578125,
|
| 1190 |
+
"learning_rate": 0.00026480000000000004,
|
| 1191 |
+
"loss": 7.6589,
|
| 1192 |
+
"step": 169
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.6302131603336423,
|
| 1196 |
+
"grad_norm": 18157.03515625,
|
| 1197 |
+
"learning_rate": 0.000264,
|
| 1198 |
+
"loss": 7.7017,
|
| 1199 |
+
"step": 170
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.633920296570899,
|
| 1203 |
+
"grad_norm": 20085.443359375,
|
| 1204 |
+
"learning_rate": 0.0002632,
|
| 1205 |
+
"loss": 7.7293,
|
| 1206 |
+
"step": 171
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.6376274328081557,
|
| 1210 |
+
"grad_norm": 26864.5390625,
|
| 1211 |
+
"learning_rate": 0.00026240000000000004,
|
| 1212 |
+
"loss": 7.5853,
|
| 1213 |
+
"step": 172
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.6413345690454124,
|
| 1217 |
+
"grad_norm": 21249.70703125,
|
| 1218 |
+
"learning_rate": 0.0002616,
|
| 1219 |
+
"loss": 7.7276,
|
| 1220 |
+
"step": 173
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.6450417052826691,
|
| 1224 |
+
"grad_norm": 17884.49609375,
|
| 1225 |
+
"learning_rate": 0.0002608,
|
| 1226 |
+
"loss": 7.7034,
|
| 1227 |
+
"step": 174
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.6487488415199258,
|
| 1231 |
+
"grad_norm": 19097.380859375,
|
| 1232 |
+
"learning_rate": 0.00026000000000000003,
|
| 1233 |
+
"loss": 7.7472,
|
| 1234 |
+
"step": 175
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.6524559777571826,
|
| 1238 |
+
"grad_norm": 21432.216796875,
|
| 1239 |
+
"learning_rate": 0.0002592,
|
| 1240 |
+
"loss": 7.7052,
|
| 1241 |
+
"step": 176
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.6561631139944393,
|
| 1245 |
+
"grad_norm": 17022.677734375,
|
| 1246 |
+
"learning_rate": 0.00025840000000000005,
|
| 1247 |
+
"loss": 7.7127,
|
| 1248 |
+
"step": 177
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.659870250231696,
|
| 1252 |
+
"grad_norm": 21216.20703125,
|
| 1253 |
+
"learning_rate": 0.00025760000000000003,
|
| 1254 |
+
"loss": 7.5911,
|
| 1255 |
+
"step": 178
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.6635773864689527,
|
| 1259 |
+
"grad_norm": 21638.240234375,
|
| 1260 |
+
"learning_rate": 0.0002568,
|
| 1261 |
+
"loss": 7.5969,
|
| 1262 |
+
"step": 179
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.6672845227062094,
|
| 1266 |
+
"grad_norm": 27894.361328125,
|
| 1267 |
+
"learning_rate": 0.00025600000000000004,
|
| 1268 |
+
"loss": 7.6331,
|
| 1269 |
+
"step": 180
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.6709916589434661,
|
| 1273 |
+
"grad_norm": 21034.33984375,
|
| 1274 |
+
"learning_rate": 0.0002552,
|
| 1275 |
+
"loss": 7.6371,
|
| 1276 |
+
"step": 181
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.6746987951807228,
|
| 1280 |
+
"grad_norm": 25746.513671875,
|
| 1281 |
+
"learning_rate": 0.0002544,
|
| 1282 |
+
"loss": 7.6462,
|
| 1283 |
+
"step": 182
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.6784059314179796,
|
| 1287 |
+
"grad_norm": 23690.24609375,
|
| 1288 |
+
"learning_rate": 0.00025360000000000004,
|
| 1289 |
+
"loss": 7.6662,
|
| 1290 |
+
"step": 183
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.6821130676552364,
|
| 1294 |
+
"grad_norm": 19138.052734375,
|
| 1295 |
+
"learning_rate": 0.0002528,
|
| 1296 |
+
"loss": 7.74,
|
| 1297 |
+
"step": 184
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.6858202038924931,
|
| 1301 |
+
"grad_norm": 20391.046875,
|
| 1302 |
+
"learning_rate": 0.000252,
|
| 1303 |
+
"loss": 7.7163,
|
| 1304 |
+
"step": 185
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.6895273401297498,
|
| 1308 |
+
"grad_norm": 17356.830078125,
|
| 1309 |
+
"learning_rate": 0.00025120000000000003,
|
| 1310 |
+
"loss": 7.6277,
|
| 1311 |
+
"step": 186
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.6932344763670065,
|
| 1315 |
+
"grad_norm": 27145.943359375,
|
| 1316 |
+
"learning_rate": 0.0002504,
|
| 1317 |
+
"loss": 7.7351,
|
| 1318 |
+
"step": 187
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.6969416126042632,
|
| 1322 |
+
"grad_norm": 18061.5703125,
|
| 1323 |
+
"learning_rate": 0.0002496,
|
| 1324 |
+
"loss": 7.6895,
|
| 1325 |
+
"step": 188
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.70064874884152,
|
| 1329 |
+
"grad_norm": 17943.388671875,
|
| 1330 |
+
"learning_rate": 0.00024880000000000003,
|
| 1331 |
+
"loss": 7.7073,
|
| 1332 |
+
"step": 189
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.7043558850787767,
|
| 1336 |
+
"grad_norm": 19911.068359375,
|
| 1337 |
+
"learning_rate": 0.000248,
|
| 1338 |
+
"loss": 7.7247,
|
| 1339 |
+
"step": 190
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.7080630213160334,
|
| 1343 |
+
"grad_norm": 23313.1328125,
|
| 1344 |
+
"learning_rate": 0.0002472,
|
| 1345 |
+
"loss": 7.6459,
|
| 1346 |
+
"step": 191
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.7117701575532901,
|
| 1350 |
+
"grad_norm": 18374.34375,
|
| 1351 |
+
"learning_rate": 0.0002464,
|
| 1352 |
+
"loss": 7.6853,
|
| 1353 |
+
"step": 192
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.7154772937905468,
|
| 1357 |
+
"grad_norm": 18763.783203125,
|
| 1358 |
+
"learning_rate": 0.0002456,
|
| 1359 |
+
"loss": 7.6097,
|
| 1360 |
+
"step": 193
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.7191844300278035,
|
| 1364 |
+
"grad_norm": 18051.265625,
|
| 1365 |
+
"learning_rate": 0.0002448,
|
| 1366 |
+
"loss": 7.6265,
|
| 1367 |
+
"step": 194
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.7228915662650602,
|
| 1371 |
+
"grad_norm": 21930.23828125,
|
| 1372 |
+
"learning_rate": 0.000244,
|
| 1373 |
+
"loss": 7.7064,
|
| 1374 |
+
"step": 195
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.726598702502317,
|
| 1378 |
+
"grad_norm": 21661.873046875,
|
| 1379 |
+
"learning_rate": 0.0002432,
|
| 1380 |
+
"loss": 7.7374,
|
| 1381 |
+
"step": 196
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.7303058387395737,
|
| 1385 |
+
"grad_norm": 26628.837890625,
|
| 1386 |
+
"learning_rate": 0.0002424,
|
| 1387 |
+
"loss": 7.6806,
|
| 1388 |
+
"step": 197
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.7340129749768304,
|
| 1392 |
+
"grad_norm": 24882.0234375,
|
| 1393 |
+
"learning_rate": 0.0002416,
|
| 1394 |
+
"loss": 7.6327,
|
| 1395 |
+
"step": 198
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.7377201112140871,
|
| 1399 |
+
"grad_norm": 25492.328125,
|
| 1400 |
+
"learning_rate": 0.0002408,
|
| 1401 |
+
"loss": 7.6956,
|
| 1402 |
+
"step": 199
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.7414272474513438,
|
| 1406 |
+
"grad_norm": 27734.201171875,
|
| 1407 |
+
"learning_rate": 0.00024,
|
| 1408 |
+
"loss": 7.6169,
|
| 1409 |
+
"step": 200
|
| 1410 |
+
}
|
| 1411 |
+
],
|
| 1412 |
+
"logging_steps": 1,
|
| 1413 |
+
"max_steps": 500,
|
| 1414 |
+
"num_input_tokens_seen": 0,
|
| 1415 |
+
"num_train_epochs": 2,
|
| 1416 |
+
"save_steps": 200,
|
| 1417 |
+
"stateful_callbacks": {
|
| 1418 |
+
"TrainerControl": {
|
| 1419 |
+
"args": {
|
| 1420 |
+
"should_epoch_stop": false,
|
| 1421 |
+
"should_evaluate": false,
|
| 1422 |
+
"should_log": false,
|
| 1423 |
+
"should_save": true,
|
| 1424 |
+
"should_training_stop": false
|
| 1425 |
+
},
|
| 1426 |
+
"attributes": {}
|
| 1427 |
+
}
|
| 1428 |
+
},
|
| 1429 |
+
"total_flos": 1.78927800680448e+16,
|
| 1430 |
+
"train_batch_size": 6,
|
| 1431 |
+
"trial_name": null,
|
| 1432 |
+
"trial_params": null
|
| 1433 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2b0885aabf5062bce5be067bc22e23dfeed6563307742b230545b8e8cbab6b6c
|
| 3 |
+
size 5112
|