root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-12-25 10:54:36 +00:00
Code Issues Actions 9 Packages Projects Releases Wiki Activity

falcon : write file type

Browse Source
This commit is contained in:
Georgi Gerganov 2023-08-24 19:58:30 +03:00
parent 38b16dfca6
commit 01f2224682
No known key found for this signature in database
GPG Key ID: 449E073F9DC10735
2 changed files with 2 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
convert-falcon-hf-to-gguf.py
View File

@ -107,6 +107,7 @@ if "n_head_kv" in hparams:
else: else:
gguf_writer.add_head_count_kv(1) gguf_writer.add_head_count_kv(1)
gguf_writer.add_layer_norm_eps(hparams["layer_norm_epsilon"]) gguf_writer.add_layer_norm_eps(hparams["layer_norm_epsilon"])
gguf_writer.add_file_type(ftype)
# TOKENIZATION # TOKENIZATION

2
scripts/run-all-ppl.sh
View File

@ -1,7 +1,7 @@
#!/bin/bash #!/bin/bash
qnt=(f16 q8_0 q6_k q5_k q5_1 q5_0 q4_k q4_1 q4_0 q3_k q2_k) qnt=(f16 q8_0 q6_k q5_k q5_1 q5_0 q4_k q4_1 q4_0 q3_k q2_k)
args="--no-mmap -ngl 999 -t 8" args="-ngl 999 -t 8"
if [ -z "$1" ]; then if [ -z "$1" ]; then
echo "usage: $0 <model> [qnt] [args]" echo "usage: $0 <model> [qnt] [args]"