[Bug] InternLM2MLP.forward() missing 1 required positional argument: 'im_mask'

[jiangjingz]

Checklist

• 1. I have searched related issues but cannot get the expected help.
• 2. The bug has not been fixed in the latest version.

Describe the bug

尝试对internlm_xcomposer_vl_7b做量化时，report以下错误：
TypeError: InternLM2MLP.forward() missing 1 required positional argument: 'im_mask'

Reproduction

lmdeploy lite auto_awq
$HF_MODEL
--calib-dataset 'ptb'
--calib-samples 128
--calib-seqlen 2048
--batch-size 1
--w-bits 4
--w-group-size 128
--search-scale False
--work-dir $WORK_DIR

Environment

python 3.10
torch 2.0.1
lmdeploy 0.4.2

Error traceback

Traceback (most recent call last):
  File "/data/manxue.jj/anaconda3/envs/py310_torch201/bin/lmdeploy", line 8, in <module>
    sys.exit(run())
  File "/data/manxue.jj/anaconda3/envs/py310_torch201/lib/python3.10/site-packages/lmdeploy/cli/entrypoint.py", line 37, in run
    args.run(args)
  File "/data/manxue.jj/anaconda3/envs/py310_torch201/lib/python3.10/site-packages/lmdeploy/cli/lite.py", line 137, in auto_awq
    auto_awq(**kwargs)
  File "/data/manxue.jj/anaconda3/envs/py310_torch201/lib/python3.10/site-packages/lmdeploy/lite/apis/auto_awq.py", line 96, in auto_awq
    vl_model, model, tokenizer, work_dir = calibrate(model,
  File "/data/manxue.jj/anaconda3/envs/py310_torch201/lib/python3.10/site-packages/lmdeploy/lite/apis/calibrate.py", line 235, in calibrate
    calib_ctx.calibrate(all_data)
  File "/data/manxue.jj/anaconda3/envs/py310_torch201/lib/python3.10/site-packages/lmdeploy/lite/quantization/calibration.py", line 315, in calibrate
    _ = model(data.to(self.device))
  File "/data/manxue.jj/anaconda3/envs/py310_torch201/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1501, in _call_impl
    return forward_call(*args, **kwargs)
  File "/home/manxue.jj/.cache/huggingface/modules/transformers_modules/trained/modeling_internlm2.py", line 929, in forward
    layer_outputs = decoder_layer(
  File "/data/manxue.jj/anaconda3/envs/py310_torch201/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1501, in _call_impl
    return forward_call(*args, **kwargs)
  File "/data/manxue.jj/anaconda3/envs/py310_torch201/lib/python3.10/site-packages/torch/utils/_contextlib.py", line 115, in decorate_context
    return func(*args, **kwargs)
  File "/data/manxue.jj/anaconda3/envs/py310_torch201/lib/python3.10/site-packages/lmdeploy/lite/quantization/calibration.py", line 505, in _forward
    auto_scale_block(mod, batch_kwargs[i], self.w_bits,
  File "/data/manxue.jj/anaconda3/envs/py310_torch201/lib/python3.10/site-packages/torch/utils/_contextlib.py", line 115, in decorate_context
    return func(*args, **kwargs)
  File "/data/manxue.jj/anaconda3/envs/py310_torch201/lib/python3.10/site-packages/lmdeploy/lite/quantization/calibration.py", line 407, in auto_scale_block
    _auto_get_scale(
  File "/data/manxue.jj/anaconda3/envs/py310_torch201/lib/python3.10/site-packages/lmdeploy/lite/quantization/calibration.py", line 400, in _auto_get_scale
    best_ratio = _search_module_scale(module2inspect, layers, inp.value,
  File "/data/manxue.jj/anaconda3/envs/py310_torch201/lib/python3.10/site-packages/lmdeploy/lite/quantization/calibration.py", line 352, in _search_module_scale
    org_out = block(x, **kwargs)
  File "/data/manxue.jj/anaconda3/envs/py310_torch201/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1501, in _call_impl
    return forward_call(*args, **kwargs)
TypeError: InternLM2MLP.forward() missing 1 required positional argument: 'im_mask'

[hekaijie123]

我发生了同样的问题，当我尝试给出--search-scale 和 --batch-size 参数时候就会出现这个问题。
lmdeploy lite auto_awq internlm-xcomposer2-vl-7b --work-dir fix_vit_newbalancetag-4bit 只有--work-dir 参数则不会出现这个问题。

[AllentDan]

Fixed in #1890