Daisuke675 commited on
Commit
5ab877d
·
verified ·
1 Parent(s): 2c4388a

final ScugnizzLLM-1B pretrain

Browse files
training-runs/pretrain-750m/checkpoint_last.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:44901311dcf4098101ae062348aa74ed071dc59f926bd8bd23010e438866cac9
3
  size 15758460435
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:841e1f67eee07c0199e8e9c96ba7fa8b4b9a811464eebd3f67685625f78ac5b6
3
  size 15758460435
training-runs/pretrain-750m/model_final.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6dd08b1f47b270d9639613ee70b9362f1e732c266bb9029d97c151e482dd1530
3
+ size 5252809683
training-runs/pretrain-750m/tokenizer/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
training-runs/pretrain-750m/tokenizer/tokenizer_config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "backend": "tokenizers",
4
+ "bos_token": "<|endoftext|>",
5
+ "eos_token": "<|endoftext|>",
6
+ "errors": "replace",
7
+ "is_local": false,
8
+ "local_files_only": false,
9
+ "model_max_length": 1024,
10
+ "pad_token": "<|endoftext|>",
11
+ "tokenizer_class": "GPT2Tokenizer",
12
+ "unk_token": "<|endoftext|>"
13
+ }