Remove unused parameter in AutoAWQ

models/config.yaml

@@ -20,8 +20,6 @@
   model_type: 'dollyv2'
 .*replit:
   model_type: 'replit'
-.*AWQ:
-  n_batch: 1
 .*(oasst|openassistant-|stablelm-7b-sft-v7-epoch-3):
   instruction_template: 'Open Assistant'
   skip_special_tokens: false

modules/loaders.py

@@ -135,7 +135,6 @@ loaders_and_params = OrderedDict({
         'gpu_memory',
         'auto_devices',
         'max_seq_len',
-        'n_batch',
         'no_inject_fused_attention',
         'trust_remote_code',
         'use_fast',

modules/models.py

@@ -298,7 +298,7 @@ def AutoAWQ_loader(model_name):
                 trust_remote_code=shared.args.trust_remote_code,
                 fuse_layers=not shared.args.no_inject_fused_attention,
                 max_memory=get_max_memory_dict(),
-                batch_size=shared.args.n_batch,
+                batch_size=1,
                 safetensors=any(model_dir.glob('*.safetensors')),
             )