PrunaAI

/

cognitivecomputations-dolphin-2_6-phi-2-QUANTO-int8bit-smashed

like 0

• "transformers.activations.NewGELUActivation",
• "transformers_modules.cognitivecomputations.dolphin-2_6-phi-2.8f9ec20eb43118517758278024f6ed911c79ee9f.modeling_phi.SelfAttention",
• "transformers_modules.cognitivecomputations.dolphin-2_6-phi-2.8f9ec20eb43118517758278024f6ed911c79ee9f.modeling_phi.MHA",
• "collections.OrderedDict",
• "torch.float16",
• "transformers_modules.cognitivecomputations.dolphin-2_6-phi-2.8f9ec20eb43118517758278024f6ed911c79ee9f.modeling_phi.Embedding",
• "torch._utils._rebuild_parameter",
• "__builtin__.set",
• "transformers_modules.cognitivecomputations.dolphin-2_6-phi-2.8f9ec20eb43118517758278024f6ed911c79ee9f.configuration_phi.PhiConfig",
• "transformers_modules.cognitivecomputations.dolphin-2_6-phi-2.8f9ec20eb43118517758278024f6ed911c79ee9f.modeling_phi.CausalLMLoss",
• "torch.device",
• "torch.int8",
• "quanto.nn.qlinear.QLinear",
• "torch.HalfStorage",
• "transformers_modules.cognitivecomputations.dolphin-2_6-phi-2.8f9ec20eb43118517758278024f6ed911c79ee9f.modeling_phi.PhiForCausalLM",
• "transformers_modules.cognitivecomputations.dolphin-2_6-phi-2.8f9ec20eb43118517758278024f6ed911c79ee9f.modeling_phi.CrossAttention",
• "transformers.generation.configuration_utils.GenerationConfig",
• "transformers_modules.cognitivecomputations.dolphin-2_6-phi-2.8f9ec20eb43118517758278024f6ed911c79ee9f.modeling_phi.ParallelBlock",
• "quanto.tensor.qtype.qtype",
• "transformers_modules.cognitivecomputations.dolphin-2_6-phi-2.8f9ec20eb43118517758278024f6ed911c79ee9f.modeling_phi.CausalLMHead",
• "transformers_modules.cognitivecomputations.dolphin-2_6-phi-2.8f9ec20eb43118517758278024f6ed911c79ee9f.modeling_phi.MLP",
• "transformers_modules.cognitivecomputations.dolphin-2_6-phi-2.8f9ec20eb43118517758278024f6ed911c79ee9f.modeling_phi.PhiModel",
• "torch._utils._rebuild_tensor_v2",
• "torch.nn.modules.dropout.Dropout",
• "torch.nn.modules.container.ModuleList",
• "torch.nn.modules.normalization.LayerNorm",
• "torch.nn.modules.sparse.Embedding",
• "torch.nn.modules.loss.CrossEntropyLoss",
• "transformers_modules.cognitivecomputations.dolphin-2_6-phi-2.8f9ec20eb43118517758278024f6ed911c79ee9f.modeling_phi.RotaryEmbedding",
• "torch.FloatStorage"

How to fix it?