tocmo0nlord/axolotl
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

Feat: Add support for upstream FA2 (#626)

Browse Source 
* Feat: Add support for upstream FA2

* chore: add is_falcon_derived_model: true to examples

* chore: add config to readme for documentation

* feat: add extra model types

* fix: remove old falcon flash patch

* chore: pin transformers and accelerate
This commit is contained in:
NanoCode012
2023-09-26 22:53:28 +09:00
committed by GitHub
parent 5e5296a77c
commit 19a600a8b8
8 changed files with 31 additions and 117 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
README.md
View File

@@ -408,6 +408,10 @@ tokenizer_legacy:
# this is reported to improve training speed on some models
resize_token_embeddings_to_32x:
# used to identify if the model is falcon/llama based
is_falcon_derived_model:
is_llama_derived_model:
# whether you are training a 4-bit GPTQ quantized model
gptq: true
gptq_groupsize: 128 # group size