PrunaAI

/

bigcode-starcoder2-7b-QUANTO-int8bit-smashed

like 0

• "transformers.models.starcoder2.modeling_starcoder2.Starcoder2RotaryEmbedding",
• "torch.nn.modules.normalization.LayerNorm",
• "transformers.generation.configuration_utils.GenerationConfig",
• "quanto.tensor.qtype.qtype",
• "torch._utils._rebuild_parameter",
• "collections.OrderedDict",
• "transformers.activations.PytorchGELUTanh",
• "transformers.models.starcoder2.modeling_starcoder2.Starcoder2MLP",
• "torch._utils._rebuild_tensor_v2",
• "transformers.models.starcoder2.configuration_starcoder2.Starcoder2Config",
• "torch.int8",
• "torch.nn.modules.sparse.Embedding",
• "transformers.models.starcoder2.modeling_starcoder2.Starcoder2ForCausalLM",
• "torch.FloatStorage",
• "transformers.models.starcoder2.modeling_starcoder2.Starcoder2SdpaAttention",
• "torch.device",
• "torch.nn.modules.container.ModuleList",
• "transformers.models.starcoder2.modeling_starcoder2.Starcoder2DecoderLayer",
• "__builtin__.set",
• "torch.bfloat16",
• "transformers.models.starcoder2.modeling_starcoder2.Starcoder2Model",
• "torch.BFloat16Storage",
• "quanto.nn.qlinear.QLinear"

How to fix it?