Skip to content

convert.py : try to determine n_ctx automatically for CodeLlama #2770

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

Merged
merged 1 commit into from
Aug 24, 2023
Merged

convert.py : try to determine n_ctx automatically for CodeLlama #2770

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
12 changes: 11 additions & 1 deletion convert.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -200,13 +200,23 @@ def loadOriginalParamsJson(model: 'LazyModel', config_path: 'Path') -> 'Params':
n_embd = config["dim"]
n_layer = config["n_layers"]
n_mult = config["multiple_of"]
n_ctx = 2048 if config["norm_eps"] == 1e-06 else 4096 # hack to determine LLaMA v1 vs v2
n_ff = -1
n_head = config["n_heads"]
n_head_kv = config["n_kv_heads"] if "n_kv_heads" in config else n_head
f_norm_eps = config["norm_eps"]
f_rope_freq_base = config["rope_theta"] if "rope_theta" in config else None

# hack to determine LLaMA v1 vs v2 vs CodeLlama
if f_rope_freq_base and f_rope_freq_base == 1000000:
# CodeLlama
n_ctx = 16384
elif config["norm_eps"] == 1e-05:
# LLaMA v2
n_ctx = 4096
else:
# LLaMA v1
n_ctx = 2048

if n_vocab == -1:
n_vocab = model["tok_embeddings.weight"].shape[0]

Expand Down