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

fix: gemma3_text model loading vision config (#3354)

Browse Source 
* fix: gemma3-text mode loading vision config

* fix: improve defaults to use lora kernels
This commit is contained in:
NanoCode012
2026-01-13 21:49:23 +07:00
committed by GitHub
parent 258ce8d4fa
commit 359b7ad85e
6 changed files with 24 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
examples/gemma3/gemma-3-4b-qlora.yml
View File

@@ -2,6 +2,7 @@ base_model: google/gemma-3-4b-it
# Need to set else transformers tries to load vision too
model_type: Gemma3ForCausalLM
cls_model_config: Gemma3TextConfig
load_in_4bit: true
@@ -32,8 +33,8 @@ sample_packing: true
lora_r: 32
lora_alpha: 16
lora_dropout: 0.05
lora_target_modules: 'model.language_model.layers.[\d]+.(mlp|cross_attn|self_attn).(up|down|gate|q|k|v|o)_proj'
lora_dropout: 0
lora_target_linear: true
wandb_project:
wandb_entity: