pilancilab

/

CALDERA-compressed-models

like 0

Follow

Pilanci Lab 6

• "torch.nn.modules.container.ModuleList",
• "transformers.models.llama.modeling_llama.LlamaDecoderLayer",
• "lib.codebook.latticee8_padded12_rvq4bit.E8P12RVQ4B_codebook",
• "torch.FloatStorage",
• "caldera.decomposition.quantized_layer.CalderaQuantizedLinear",
• "torch.nn.modules.linear.Linear",
• "transformers.models.llama.modeling_llama.LlamaRMSNorm",
• "torch._utils._rebuild_tensor_v2",
• "torch.BFloat16Storage",
• "torch.nn.modules.activation.SiLU",
• "torch._utils._rebuild_parameter",
• "transformers.models.llama.modeling_llama.LlamaForCausalLM",
• "torch.BoolStorage",
• "torch.nn.modules.sparse.Embedding",
• "transformers.models.llama.modeling_llama.LlamaMLP",
• "lib.codebook.latticee8_padded12.E8P12_codebook",
• "torch.LongStorage",
• "caldera.decomposition.quantized_layer.LatticeQuantizedParameter",
• "torch.int64",
• "transformers.models.llama.configuration_llama.LlamaConfig",
• "transformers.generation.configuration_utils.GenerationConfig",
• "collections.OrderedDict",
• "__builtin__.set",
• "torch.HalfStorage",
• "transformers.models.llama.modeling_llama.LlamaRotaryEmbedding",
• "transformers.models.llama.modeling_llama.LlamaSdpaAttention",
• "transformers.models.llama.modeling_llama.LlamaModel",
• "torch.float16",
• "torch.IntStorage"

How to fix it?

• "lib.codebook.latticee8_padded12_rvq4bit.E8P12RVQ4B_codebook",
• "torch.nn.modules.activation.SiLU",
• "transformers.models.llama.modeling_llama.LlamaRotaryEmbedding",
• "torch._utils._rebuild_parameter",
• "transformers.generation.configuration_utils.GenerationConfig",
• "transformers.models.llama.modeling_llama.LlamaSdpaAttention",
• "transformers.models.llama.modeling_llama.LlamaRMSNorm",
• "caldera.decomposition.quantized_layer.CalderaQuantizedLinear",
• "transformers.models.llama.modeling_llama.LlamaModel",
• "lib.codebook.latticee8_padded12.E8P12_codebook",
• "caldera.decomposition.quantized_layer.LatticeQuantizedParameter",
• "__builtin__.set",
• "transformers.models.llama.modeling_llama.LlamaMLP",
• "torch.HalfStorage",
• "collections.OrderedDict",
• "torch.LongStorage",
• "torch.nn.modules.sparse.Embedding",
• "torch.FloatStorage",
• "transformers.models.llama.modeling_llama.LlamaForCausalLM",
• "transformers.models.llama.configuration_llama.LlamaConfig",
• "torch.nn.modules.container.ModuleList",
• "torch._utils._rebuild_tensor_v2",
• "torch.int64",
• "torch.BFloat16Storage",
• "torch.nn.modules.linear.Linear",
• "torch.BoolStorage",
• "transformers.models.llama.modeling_llama.LlamaDecoderLayer",
• "torch.float16",
• "torch.IntStorage"

How to fix it?