PrunaAI

/

jslin09-bloom-560m-finetuned-fraud-QUANTO-float8bit-smashed

like 0

Follow

Pruna AI 152

• "torch.float8_e4m3fn",
• "torch.device",
• "torch._utils._rebuild_parameter",
• "quanto.nn.qlinear.QLinear",
• "transformers.models.bloom.configuration_bloom.BloomConfig",
• "transformers.models.bloom.modeling_bloom.BloomForCausalLM",
• "torch.nn.modules.container.ModuleList",
• "torch.FloatStorage",
• "quanto.tensor.qtype.qtype",
• "transformers.models.bloom.modeling_bloom.BloomBlock",
• "torch._utils._rebuild_tensor_v2",
• "collections.OrderedDict",
• "transformers.models.bloom.modeling_bloom.BloomGelu",
• "torch.nn.modules.normalization.LayerNorm",
• "transformers.models.bloom.modeling_bloom.BloomAttention",
• "__builtin__.set",
• "transformers.generation.configuration_utils.GenerationConfig",
• "torch.nn.modules.sparse.Embedding",
• "transformers.models.bloom.modeling_bloom.BloomModel",
• "torch.float32",
• "torch.nn.modules.dropout.Dropout",
• "transformers.models.bloom.modeling_bloom.BloomMLP"

How to fix it?