Upload HfMoondream

config.py

@@ -38,7 +38,7 @@ class RegionConfig:
     size_feat_dim: int = 512
     size_out_dim: int = 2048
     inner_dim: int = 8192
-    group_size: Optional[int] = 128
+    group_size: Optional[int] = None
 
 @dataclass(frozen=True)
 class TokenizerConfig:

layers.py

@@ -7,8 +7,6 @@ from typing import Literal
 from torchao import quantize_
 from torchao.quantization import int4_weight_only
 
-from .packing import dequantize_tensor
-
 
 def gelu_approx(x):
     return F.gelu(x, approximate="tanh")
@@ -24,6 +22,15 @@ def linear(x: torch.Tensor, w: LinearWeights) -> torch.Tensor:
     return F.linear(x, w.weight, w.bias)
 
 
+def dequantize_tensor(W_q, scale, zero, orig_shape, dtype=torch.bfloat16):
+    _step = W_q.shape[0]
+    W_r = torch.empty([2 * _step, W_q.shape[1]], dtype=dtype, device=W_q.device)
+    W_r[:_step] = (W_q & 0b11110000) >> 4
+    W_r[_step:] = W_q & 0b00001111
+    W_r.sub_(zero).mul_(scale)
+    return W_r.reshape(orig_shape)
+
+
 class QuantizedLinear(nn.Module):
     def __init__(
         self,

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b839cdbd6716eef6242536929c05243d58af49929a12c198d3913caa05c7c3ee
-size 2032380848
+oid sha256:de670ae0501f95abc4b15669d80da1551cb5ec66ce80bbae0fb290c212d5eb95
+size 2119003880