PrunaAI

/

princeton-nlp-gemma-2-9b-it-SimPO-QUANTO-float8bit-smashed

like 0

Follow

Pruna AI 118

• "torch.bfloat16",
• "torch.BFloat16Storage",
• "transformers.models.gemma2.modeling_gemma2.Gemma2DecoderLayer",
• "quanto.tensor.qtype.qtype",
• "transformers.activations.PytorchGELUTanh",
• "transformers.models.gemma2.modeling_gemma2.Gemma2SdpaAttention",
• "torch.nn.modules.sparse.Embedding",
• "torch.float8_e4m3fn",
• "__builtin__.set",
• "torch._utils._rebuild_parameter",
• "transformers.models.gemma2.modeling_gemma2.Gemma2RotaryEmbedding",
• "torch.nn.modules.container.ModuleList",
• "transformers.generation.configuration_utils.GenerationConfig",
• "transformers.models.gemma2.modeling_gemma2.Gemma2MLP",
• "torch.FloatStorage",
• "torch.device",
• "transformers.models.gemma2.modeling_gemma2.Gemma2ForCausalLM",
• "transformers.models.gemma2.modeling_gemma2.Gemma2RMSNorm",
• "collections.OrderedDict",
• "torch._utils._rebuild_tensor_v2",
• "transformers.models.gemma2.modeling_gemma2.Gemma2Model",
• "quanto.nn.qlinear.QLinear",
• "transformers.models.gemma2.configuration_gemma2.Gemma2Config"

How to fix it?