Downloads · 30 days
1.6K
17% of all-time downloads
yujiepan/dbrx-tiny-random
dbrx-tiny-random is a text generation model from yujiepan. Use it when you need the model to write or continue text. It is set up for transformers.
This model is randomly initialized, using the config from databricks/dbrx-instruct but with smaller size. Note the model is in float16.
Downloads · 30 days
1.6K
17% of all-time downloads
All-time downloads
9.2K
Public
Parameters
806K
1.6 MB on disk
Likes
0
Public
Click a slice to open those files.
.json5.9 MB · 70%
From the Hugging Face model README
This model is randomly initialized, using the config from databricks/dbrx-instruct but with smaller size. Note the model is in float16.
Codes:
import transformers
import torch
import os
from huggingface_hub import create_repo, upload_folder
source_model_id = 'databricks/dbrx-instruct'
save_path = '/tmp/yujiepan/dbrx-tiny-random'
repo_id = 'yujiepan/dbrx-tiny-random'
config = transformers.AutoConfig.from_pretrained(
source_model_id, trust_remote_code=True)
config.attn_config.kv_n_heads = 2
config.d_model = 4
config.ffn_config.ffn_hidden_size = 8
config.n_heads = 4
config.n_layers = 2
model = transformers.AutoModelForCausalLM.from_config(
config, trust_remote_code=True)
model = model.half()
model.save_pretrained(save_path)
tokenizer = transformers.AutoTokenizer.from_pretrained(
source_model_id, trust_remote_code=True)
tokenizer.save_pretrained(save_path)
result = transformers.pipelines.pipeline(
'text-generation',
model=model.float(), tokenizer=tokenizer)('Hello')
print(result)
os.system(f'ls -alh {save_path}')
create_repo(repo_id, exist_ok=True)
upload_folder(repo_id=repo_id, folder_path=save_path)