vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Fix model conversion script

Browse source
This commit is contained in:
Galunid 2023-10-21 11:12:01 +02:00
parent f1dd430f67
commit 4fbce39089
1 changed files with 1 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

2
convert-stablelm-hf-to-gguf.py
View file

@ -85,7 +85,7 @@ gguf_writer.add_context_length(hparams["max_position_embeddings"])
gguf_writer.add_embedding_length(hparams["hidden_size"])
gguf_writer.add_block_count(block_count)
gguf_writer.add_feed_forward_length(hparams["intermediate_size"])
gguf_writer.add_rope_dimension_count(int(hparams["rope_pct"]*hparams["hidden_size"]))
gguf_writer.add_rope_dimension_count(int(hparams["rope_pct"]*(hparams["hidden_size"] // hparams["num_attention_heads"])))
gguf_writer.add_head_count(hparams["num_attention_heads"])
gguf_writer.add_parallel_residual(hparams["use_parallel_residual"] if "use_parallel_residual" in hparams else True)
gguf_writer.add_layer_norm_eps(1e-5)