PrunaAI

/

ricepaper-vi-gemma-2b-RAG-QUANTO-int4bit-smashed

like 0

Follow

Pruna AI 146

• "torch._utils._rebuild_parameter",
• "torch.FloatStorage",
• "torch.BFloat16Storage",
• "collections.OrderedDict",
• "torch.int8",
• "transformers.models.gemma.modeling_gemma.GemmaRMSNorm",
• "__builtin__.set",
• "quanto.nn.qlinear.QLinear",
• "transformers.models.gemma.modeling_gemma.GemmaSdpaAttention",
• "transformers.generation.configuration_utils.GenerationConfig",
• "transformers.models.gemma.modeling_gemma.GemmaForCausalLM",
• "transformers.models.gemma.modeling_gemma.GemmaMLP",
• "torch.device",
• "torch.nn.modules.container.ModuleList",
• "quanto.tensor.qtype.qtype",
• "transformers.models.gemma.modeling_gemma.GemmaRotaryEmbedding",
• "transformers.activations.PytorchGELUTanh",
• "transformers.models.gemma.configuration_gemma.GemmaConfig",
• "torch.nn.modules.sparse.Embedding",
• "torch._utils._rebuild_tensor_v2",
• "transformers.models.gemma.modeling_gemma.GemmaModel",
• "transformers.models.gemma.modeling_gemma.GemmaDecoderLayer",
• "torch.bfloat16"

How to fix it?