CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-11-03 09:22:01 +00:00
Code Issues Packages Projects Releases Wiki Activity

convert-falcon-hf-to-gguf.py : add tensor data layout

Browse Source
This commit is contained in:
klosax
2023-08-19 12:05:11 +02:00
committed by GitHub
parent 2c8055b65b
commit b3a7a2b486
1 changed files with 2 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
convert-falcon-hf-to-gguf.py
View File

@@ -95,7 +95,8 @@ print("gguf: get model metadata")
block_count = hparams["n_layer"]
gguf_writer.add_name(last_dir)
gguf_writer.add_context_length(2048)
gguf_writer.add_context_length(2048) # not in config.json
gguf_writer.add_tensor_data_layout("jploski") # qkv tensor transform
gguf_writer.add_embedding_length(hparams["hidden_size"])
gguf_writer.add_feed_forward_length(4 * hparams["hidden_size"])
gguf_writer.add_block_count(block_count)