kloodia/M8x7-adapter-pretrain-vision

These are weights for a version of mistralai/Mixtral-8x7B-Instruct-v0.1 finetuned for multimodal applications.

Modalities

• CLIPVisionModality (use <image> in text and provide images, encoded as 576 tokens)

Usage

GitHub: https://github.com/sshh12/multi_token (includes training scripts and basic inference server)

Dataset

./out (558128 examples)

{'id': '004539375', 'images': ['/data/llava_pretrain_data/images/00453/004539375.jpg'], 'messages': [{'content': 'Render a clear and concise summary of the photo.\n<image>', 'role': 'user'}, {'content': 'select luxury furniture 3 - inch gel memory foam mattress topper', 'role': 'assistant'}]}

Training Device(s)

name, pci.bus_id, vbios_version
NVIDIA GeForce RTX 3090, 00000000:B3:00.0, 94.02.42.00.B4

Model

MistralLMMForCausalLM.model =

PeftModelForCausalLM(
  (base_model): LoraModel(
    (model): MistralLMMForCausalLM(
      (model): MistralLMMModel(
        (embed_tokens): Embedding(32000, 4096)
        (layers): ModuleList(
          (0-31): 32 x MistralDecoderLayer(
            (self_attn): MistralAttention(
              (q_proj): lora.Linear(
                (base_layer): Linear(in_features=4096, out_features=4096, bias=False)
                (lora_dropout): ModuleDict(
                  (default): Dropout(p=0.05, inplace=False)
                )
                (lora_A): ModuleDict(
                  (default): Linear(in_features=4096, out_features=64, bias=False)
                )
                (lora_B): ModuleDict(
                  (default): Linear(in_features=64, out_features=4096, bias=False)
                )
                (lora_embedding_A): ParameterDict()
                (lora_embedding_B): ParameterDict()
              )
              (k_proj): lora.Linear(
                (base_layer): Linear(in_features=4096, out_features=1024, bias=False)
                (lora_dropout): ModuleDict(
                  (default): Dropout(p=0.05, inplace=False)
                )
                (lora_A): ModuleDict(
                  (default): Linear(in_features=4096, out_features=64, bias=False)
                )
                (lora_B): ModuleDict(
                  (default): Linear(in_features=64, out_features=1024, bias=False)
                )
                (lora_embedding_A): ParameterDict()
                (lora_embedding_B): ParameterDict()
              )
              (v_proj): lora.Linear(
                (base_layer): Linear(in_features=4096, out_features=1024, bias=False)
                (lora_dropout): ModuleDict(
                  (default): Dropout(p=0.05, inplace=False)
                )
                (lora_A): ModuleDict(
                  (default): Linear(in_features=4096, out_features=64, bias=False)
                )
                (lora_B): ModuleDict(
                  (default): Linear(in_features=64, out_features=1024, bias=False)
                )
                (lora_embedding_A): ParameterDict()
                (lora_embedding_B): ParameterDict()
              )
              (o_proj): lora.Linear(
                (base_layer): Linear(in_features=4096, out_features=4096, bias=False)
                (lora_dropout): ModuleDict(
                  (default): Dropout(p=0.05, inplace=False)
                )
                (lora_A): ModuleDict(
                  (default): Linear(in_features=4096, out_features=64, bias=False)
                )
                (lora_B): ModuleDict(
                  (default): Linear(in_features=64, out_features=4096, bias=False)
                )
                (lora_embedding_A): ParameterDict()
                (lora_embedding_B): ParameterDict()
              )
              (rotary_emb): MistralRotaryEmbedding()
            )
            (mlp): MistralMLP(
              (gate_proj): lora.Linear(
                (base_layer): Linear(in_features=4096, out_features=14336, bias=False)
                (lora_dropout): ModuleDict(
                  (default): Dropout(p=0.05, inplace=False)
                )
                (lora_A): ModuleDict(
                  (default): Linear(in_features=4096, out_features=64, bias=False)
                )
                (lora_B): ModuleDict(
                  (default): Linear(in_features=64, out_features=14336, bias=False)
                )
                (lora_embedding_A): ParameterDict()
                (lora_embedding_B): ParameterDict()
              )
              (up_proj): lora.Linear(
                (base_layer): Linear(in_features=4096, out_features=14336, bias=False)
                (lora_dropout): ModuleDict(
                  (default): Dropout(p=0.05, inplace=False)
                )
                (lora_A): ModuleDict(
                  (default): Linear(in_features=4096, out_features=64, bias=False)
                )
                (lora_B): ModuleDict(
                  (default): Linear(in_features=64, out_features=14336, bias=False)
                )
                (lora_embedding_A): ParameterDict()
                (lora_embedding_B): ParameterDict()
              )
              (down_proj): lora.Linear(
                (base_layer): Linear(in_features=14336, out_features=4096, bias=False)
                (lora_dropout): ModuleDict(
                  (default): Dropout(p=0.05, inplace=False)
                )
                (lora_A): ModuleDict(
                  (default): Linear(in_features=14336, out_features=64, bias=False)
                )
                (lora_B): ModuleDict(
                  (default): Linear(in_features=64, out_features=4096, bias=False)
                )
                (lora_embedding_A): ParameterDict()
                (lora_embedding_B): ParameterDict()
              )
              (act_fn): SiLU()
            )
            (input_layernorm): MistralRMSNorm()
            (post_attention_layernorm): MistralRMSNorm()
          )
        )
        (norm): MistralRMSNorm()
        (vision_clip_lmm_projector): Sequential(
          (0): Linear(in_features=1024, out_features=4096, bias=True)
          (1): GELU(approximate='none')
          (2): Linear(in_features=4096, out_features=4096, bias=True)
        )
      )
      (lm_head): Linear(in_features=4096, out_features=32000, bias=False)
    )
  )
)

Framework versions

• PEFT 0.10.0