PrunaAI

/

cognitivecomputations-dolphin-2.9.3-qwen2-1.5b-QUANTO-int4bit-smashed

like 0

• "torch.BFloat16Storage",
• "__builtin__.set",
• "torch._utils._rebuild_tensor_v2",
• "torch.nn.modules.container.ModuleList",
• "collections.OrderedDict",
• "torch.bfloat16",
• "transformers.generation.configuration_utils.GenerationConfig",
• "transformers.models.qwen2.modeling_qwen2.Qwen2DecoderLayer",
• "transformers.models.qwen2.modeling_qwen2.Qwen2MLP",
• "transformers.models.qwen2.modeling_qwen2.Qwen2RMSNorm",
• "torch.device",
• "torch.FloatStorage",
• "torch.nn.modules.sparse.Embedding",
• "torch.nn.modules.activation.SiLU",
• "quanto.nn.qlinear.QLinear",
• "torch._utils._rebuild_parameter",
• "transformers.models.qwen2.configuration_qwen2.Qwen2Config",
• "transformers.models.qwen2.modeling_qwen2.Qwen2RotaryEmbedding",
• "quanto.tensor.qtype.qtype",
• "transformers.models.qwen2.modeling_qwen2.Qwen2Model",
• "torch.int8",
• "transformers.models.qwen2.modeling_qwen2.Qwen2SdpaAttention",
• "transformers.models.qwen2.modeling_qwen2.Qwen2ForCausalLM"

How to fix it?