初始化项目,由ModelHub XC社区提供模型

Model: pszemraj/pythia-31m-goodwiki-deduped-2048-scratch
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-04-21 07:15:55 +08:00
commit 4e32eb53be
9 changed files with 100746 additions and 0 deletions

9
tokenizer_config.json Normal file
View File

@@ -0,0 +1,9 @@
{
"add_prefix_space": false,
"bos_token": "<|endoftext|>",
"clean_up_tokenization_spaces": true,
"eos_token": "<|endoftext|>",
"model_max_length": 1000000000000000019884624838656,
"tokenizer_class": "GPTNeoXTokenizer",
"unk_token": "<|endoftext|>"
}