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

convert.py : add rope_freq_base when converting CodeLlama from an HF model #2773

Merged
merged 1 commit into from
Aug 25, 2023
Merged

convert.py : add rope_freq_base when converting CodeLlama from an HF model #2773

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
34 changes: 18 additions & 16 deletions convert.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -160,13 +160,14 @@ def guessed(model: 'LazyModel') -> 'Params':
def loadHFTransformerJson(model: 'LazyModel', config_path: 'Path') -> 'Params':
config = json.load(open(config_path))

n_vocab = config["vocab_size"]
n_embd = config["hidden_size"]
n_layer = config["num_hidden_layers"]
n_ff = config["intermediate_size"]
n_head = config["num_attention_heads"]
n_head_kv = config["num_key_value_heads"] if "num_key_value_heads" in config else n_head
f_norm_eps = config["rms_norm_eps"]
n_vocab = config["vocab_size"]
n_embd = config["hidden_size"]
n_layer = config["num_hidden_layers"]
n_ff = config["intermediate_size"]
n_head = config["num_attention_heads"]
n_head_kv = config["num_key_value_heads"] if "num_key_value_heads" in config else n_head
f_norm_eps = config["rms_norm_eps"]
f_rope_freq_base = config["rope_theta"] if "rope_theta" in config else None

n_mult = Params.find_n_mult(n_ff, n_embd)

Expand All @@ -179,15 +180,16 @@ def loadHFTransformerJson(model: 'LazyModel', config_path: 'Path') -> 'Params':
"Suggestion: provide 'config.json' of the model in the same directory containing model files.")

return Params(
n_vocab = n_vocab,
n_embd = n_embd,
n_mult = n_mult,
n_layer = n_layer,
n_ctx = n_ctx,
n_ff = n_ff,
n_head = n_head,
n_head_kv = n_head_kv,
f_norm_eps = f_norm_eps,
n_vocab = n_vocab,
n_embd = n_embd,
n_mult = n_mult,
n_layer = n_layer,
n_ctx = n_ctx,
n_ff = n_ff,
n_head = n_head,
n_head_kv = n_head_kv,
f_norm_eps = f_norm_eps,
f_rope_freq_base = f_rope_freq_base,
)

# LLaMA v2 70B params.json
Expand Down