PrunaAI

/

voidful-phi-1_5_chat_128k-QUANTO-int4bit-smashed

like 0

Follow

Pruna AI 116

• "collections.OrderedDict",
• "torch.int8",
• "torch.nn.modules.sparse.Embedding",
• "transformers.models.phi.modeling_phi.PhiModel",
• "quanto.nn.qlinear.QLinear",
• "torch.device",
• "transformers.models.phi.modeling_phi.PhiDecoderLayer",
• "transformers.activations.NewGELUActivation",
• "torch.nn.modules.normalization.LayerNorm",
• "torch._utils._rebuild_tensor_v2",
• "torch.nn.modules.container.ModuleList",
• "torch.nn.modules.dropout.Dropout",
• "transformers.models.phi.modeling_phi.PhiForCausalLM",
• "torch.FloatStorage",
• "transformers.models.phi.configuration_phi.PhiConfig",
• "torch.bfloat16",
• "torch._utils._rebuild_parameter",
• "transformers.models.phi.modeling_phi.PhiMLP",
• "quanto.tensor.qtype.qtype",
• "__builtin__.set",
• "transformers.models.phi.modeling_phi.PhiDynamicNTKScalingRotaryEmbedding",
• "transformers.models.phi.modeling_phi.PhiSdpaAttention",
• "torch.BFloat16Storage",
• "transformers.generation.configuration_utils.GenerationConfig"

How to fix it?