opt¶
Model type |
|
Class |
|
Task |
|
Source |
|
Description¶
OPT causal language model.
Same architecture as GPT-2 (pre-norm LayerNorm, absolute positional embeddings, standard attention with KV cache) but with different HF weight naming convention and a position embedding offset of 2.
Usage¶
mobius build --model <MODEL_ID> output_dir/
from mobius import build
model = build("<MODEL_ID>")