Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

split up llama model loading so config can be loaded from base config and models can be loaded from a path #120

Merged
merged 3 commits into from
May 31, 2023
Merged

split up llama model loading so config can be loaded from base config and models can be loaded from a path #120

Changes from 1 commit
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
2520ecd
split up llama model loading so config can be loaded from base config…
winglian May 31, 2023
39a208c
fix up tokenizer config, isort fix
winglian May 31, 2023
e3c494c
remove unused import and update readme
winglian May 31, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 3 additions & 1 deletion src/axolotl/utils/models.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10,7 +10,7 @@
import bitsandbytes as bnb
import torch
import transformers
from transformers import AutoModelForCausalLM # noqa: F401
from transformers import AutoModelForCausalLM, LlamaConfig # noqa: F401
winglian marked this conversation as resolved.
Show resolved Hide resolved
from transformers import PreTrainedModel # noqa: F401
from transformers import AutoConfig, AutoTokenizer, BitsAndBytesConfig
winglian marked this conversation as resolved.
Show resolved Hide resolved

Expand Down Expand Up @@ -172,8 +172,10 @@ def load_model(
)
load_in_8bit = False
elif is_llama_derived_model and "LlamaForCausalLM" in globals():
config = LlamaConfig.from_pretrained(base_model_config)
model = LlamaForCausalLM.from_pretrained(
base_model,
config=config,
load_in_8bit=cfg.load_in_8bit and cfg.adapter is not None,
load_in_4bit=cfg.load_in_4bit and cfg.adapter is not None,
torch_dtype=torch_dtype,
Expand Down