diff --git a/timm/data/auto_augment.py b/timm/data/auto_augment.py
index 8907e504..121a3fc6 100644
--- a/timm/data/auto_augment.py
+++ b/timm/data/auto_augment.py
@@ -15,7 +15,7 @@ Papers:
     RandAugment: Practical automated data augmentation... - https://arxiv.org/abs/1909.13719
     AugMix: A Simple Data Processing Method to Improve Robustness and Uncertainty - https://arxiv.org/abs/1912.02781
 
-Hacked together by / Copyright 2020 Ross Wightman
+Hacked together by / Copyright 2019, Ross Wightman
 """
 import random
 import math
diff --git a/timm/data/dataset.py b/timm/data/dataset.py
index d3603a23..20b663ce 100644
--- a/timm/data/dataset.py
+++ b/timm/data/dataset.py
@@ -1,6 +1,6 @@
 """ Quick n Simple Image Folder, Tarfile based DataSet
 
-Hacked together by / Copyright 2020 Ross Wightman
+Hacked together by / Copyright 2019, Ross Wightman
 """
 import torch.utils.data as data
 import os
diff --git a/timm/data/dataset_factory.py b/timm/data/dataset_factory.py
index e86bcc29..194a597e 100644
--- a/timm/data/dataset_factory.py
+++ b/timm/data/dataset_factory.py
@@ -1,3 +1,7 @@
+""" Dataset Factory
+
+Hacked together by / Copyright 2021, Ross Wightman
+"""
 import os
 
 from torchvision.datasets import CIFAR100, CIFAR10, MNIST, QMNIST, KMNIST, FashionMNIST, ImageNet, ImageFolder
diff --git a/timm/data/loader.py b/timm/data/loader.py
index a02399a3..67d8cd83 100644
--- a/timm/data/loader.py
+++ b/timm/data/loader.py
@@ -3,7 +3,7 @@
 Prefetcher and Fast Collate inspired by NVIDIA APEX example at
 https://github.com/NVIDIA/apex/commit/d5e2bb4bdeedd27b1dfaf5bb2b24d6c000dee9be#diff-cf86c282ff7fba81fad27a559379d5bf
 
-Hacked together by / Copyright 2021 Ross Wightman
+Hacked together by / Copyright 2019, Ross Wightman
 """
 import random
 from functools import partial
diff --git a/timm/data/mixup.py b/timm/data/mixup.py
index 7e382c52..c8789a0c 100644
--- a/timm/data/mixup.py
+++ b/timm/data/mixup.py
@@ -8,7 +8,7 @@ CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Fea
 Code Reference:
 CutMix: https://github.com/clovaai/CutMix-PyTorch
 
-Hacked together by / Copyright 2020 Ross Wightman
+Hacked together by / Copyright 2019, Ross Wightman
 """
 import numpy as np
 import torch
diff --git a/timm/data/random_erasing.py b/timm/data/random_erasing.py
index 2fa63153..98108488 100644
--- a/timm/data/random_erasing.py
+++ b/timm/data/random_erasing.py
@@ -3,7 +3,7 @@
 Originally inspired by impl at https://github.com/zhunzhong07/Random-Erasing, Apache 2.0
 Copyright Zhun Zhong & Liang Zheng
 
-Hacked together by / Copyright 2020 Ross Wightman
+Hacked together by / Copyright 2019, Ross Wightman
 """
 import random
 import math
diff --git a/timm/data/transforms_factory.py b/timm/data/transforms_factory.py
index d4815d95..a5facbf5 100644
--- a/timm/data/transforms_factory.py
+++ b/timm/data/transforms_factory.py
@@ -1,7 +1,7 @@
 """ Transforms Factory
 Factory methods for building image transforms for use with TIMM (PyTorch Image Models)
 
-Hacked together by / Copyright 2020 Ross Wightman
+Hacked together by / Copyright 2019, Ross Wightman
 """
 import math
 
diff --git a/timm/models/cait.py b/timm/models/cait.py
index 69b4ba06..978db1ab 100644
--- a/timm/models/cait.py
+++ b/timm/models/cait.py
@@ -4,6 +4,7 @@ Paper: 'Going deeper with Image Transformers' - https://arxiv.org/abs/2103.17239
 
 Original code and weights from https://github.com/facebookresearch/deit, copyright below
 
+Modifications and additions for timm hacked together by / Copyright 2021, Ross Wightman
 """
 # Copyright (c) 2015-present, Facebook, Inc.
 # All rights reserved.
diff --git a/timm/models/convit.py b/timm/models/convit.py
index 6ef1da72..1a4c1810 100644
--- a/timm/models/convit.py
+++ b/timm/models/convit.py
@@ -9,6 +9,8 @@
 
 Paper link: https://arxiv.org/abs/2103.10697
 Original code: https://github.com/facebookresearch/convit, original copyright below
+
+Modifications and additions for timm hacked together by / Copyright 2021, Ross Wightman
 """
 # Copyright (c) 2015-present, Facebook, Inc.
 # All rights reserved.
diff --git a/timm/models/crossvit.py b/timm/models/crossvit.py
index ddc4f64c..acf9a47d 100644
--- a/timm/models/crossvit.py
+++ b/timm/models/crossvit.py
@@ -12,6 +12,8 @@ Paper link: https://arxiv.org/abs/2103.14899
 Original code: https://github.com/IBM/CrossViT/blob/main/models/crossvit.py
 
 NOTE: model names have been renamed from originals to represent actual input res all *_224 -> *_240 and *_384 -> *_408
+
+Modifications and additions for timm hacked together by / Copyright 2021, Ross Wightman
 """
 
 # Copyright IBM All Rights Reserved.
diff --git a/timm/models/cspnet.py b/timm/models/cspnet.py
index 8465e97d..4feb341c 100644
--- a/timm/models/cspnet.py
+++ b/timm/models/cspnet.py
@@ -409,7 +409,7 @@ class CspNet(nn.Module):
 def _create_cspnet(variant, pretrained=False, **kwargs):
     cfg_variant = variant.split('_')[0]
     # NOTE: DarkNet is one of few models with stride==1 features w/ 6 out_indices [0..5]
-    out_indices = kwargs.get('out_indices', (0, 1, 2, 3, 4, 5) if 'darknet' in variant else (0, 1, 2, 3, 4))
+    out_indices = kwargs.pop('out_indices', (0, 1, 2, 3, 4, 5) if 'darknet' in variant else (0, 1, 2, 3, 4))
     return build_model_with_cfg(
         CspNet, variant, pretrained,
         default_cfg=default_cfgs[variant],
diff --git a/timm/models/efficientnet.py b/timm/models/efficientnet.py
index 30d6fd6a..87e033e0 100644
--- a/timm/models/efficientnet.py
+++ b/timm/models/efficientnet.py
@@ -33,7 +33,7 @@ The majority of the above models (EfficientNet*, MixNet, MnasNet) and original w
 by Mingxing Tan, Quoc Le, and other members of their Google Brain team. Thanks for consistently releasing
 the models and weights open source!
 
-Hacked together by / Copyright 2021 Ross Wightman
+Hacked together by / Copyright 2019, Ross Wightman
 """
 from functools import partial
 from typing import List
@@ -718,7 +718,7 @@ def _gen_mobilenet_v2(
     round_chs_fn = partial(round_channels, multiplier=channel_multiplier)
     model_kwargs = dict(
         block_args=decode_arch_def(arch_def, depth_multiplier=depth_multiplier, fix_first_last=fix_stem_head),
-        num_features=1280 if fix_stem_head else round_chs_fn(1280),
+        num_features=1280 if fix_stem_head else max(1280, round_chs_fn(1280)),
         stem_size=32,
         fix_stem=fix_stem_head,
         round_chs_fn=round_chs_fn,
diff --git a/timm/models/efficientnet_blocks.py b/timm/models/efficientnet_blocks.py
index b1fec449..aef66295 100644
--- a/timm/models/efficientnet_blocks.py
+++ b/timm/models/efficientnet_blocks.py
@@ -1,6 +1,6 @@
 """ EfficientNet, MobileNetV3, etc Blocks
 
-Hacked together by / Copyright 2020 Ross Wightman
+Hacked together by / Copyright 2019, Ross Wightman
 """
 
 import torch
diff --git a/timm/models/efficientnet_builder.py b/timm/models/efficientnet_builder.py
index 7a05c890..8192f4e5 100644
--- a/timm/models/efficientnet_builder.py
+++ b/timm/models/efficientnet_builder.py
@@ -3,7 +3,7 @@
 Assembles EfficieNet and related network feature blocks from string definitions.
 Handles stride, dilation calculations, and selects feature extraction points.
 
-Hacked together by / Copyright 2020 Ross Wightman
+Hacked together by / Copyright 2019, Ross Wightman
 """
 
 import logging
diff --git a/timm/models/layers/__init__.py b/timm/models/layers/__init__.py
index 4831af9a..706d9dc8 100644
--- a/timm/models/layers/__init__.py
+++ b/timm/models/layers/__init__.py
@@ -21,7 +21,7 @@ from .helpers import to_ntuple, to_2tuple, to_3tuple, to_4tuple, make_divisible
 from .inplace_abn import InplaceAbn
 from .linear import Linear
 from .mixed_conv2d import MixedConv2d
-from .mlp import Mlp, GluMlp, GatedMlp
+from .mlp import Mlp, GluMlp, GatedMlp, ConvMlp
 from .non_local_attn import NonLocalAttn, BatNonLocalAttn
 from .norm import GroupNorm, LayerNorm2d
 from .norm_act import BatchNormAct2d, GroupNormAct
diff --git a/timm/models/levit.py b/timm/models/levit.py
index 9987e4ba..4f400fd1 100644
--- a/timm/models/levit.py
+++ b/timm/models/levit.py
@@ -14,7 +14,7 @@ Adapted from official impl at https://github.com/facebookresearch/LeViT, origina
 
 This version combines both conv/linear models and fixes torchscript compatibility.
 
-Modifications by/coyright Copyright 2021 Ross Wightman
+Modifications and additions for timm hacked together by / Copyright 2021, Ross Wightman
 """
 
 # Copyright (c) 2015-present, Facebook, Inc.
diff --git a/timm/models/mobilenetv3.py b/timm/models/mobilenetv3.py
index 4b9bcc09..6f991988 100644
--- a/timm/models/mobilenetv3.py
+++ b/timm/models/mobilenetv3.py
@@ -1,11 +1,10 @@
-
 """ MobileNet V3
 
 A PyTorch impl of MobileNet-V3, compatible with TF weights from official impl.
 
 Paper: Searching for MobileNetV3 - https://arxiv.org/abs/1905.02244
 
-Hacked together by / Copyright 2021 Ross Wightman
+Hacked together by / Copyright 2019, Ross Wightman
 """
 from functools import partial
 from typing import List
diff --git a/timm/models/resnet.py b/timm/models/resnet.py
index bbcae9a3..8989f300 100644
--- a/timm/models/resnet.py
+++ b/timm/models/resnet.py
@@ -4,7 +4,8 @@ This started as a copy of https://github.com/pytorch/vision 'resnet.py' (BSD-3-C
 additional dropout and dynamic global avg/max pool.
 
 ResNeXt, SE-ResNeXt, SENet, and MXNet Gluon stem/downsample variants, tiered stems added by Ross Wightman
-Copyright 2020 Ross Wightman
+
+Copyright 2019, Ross Wightman
 """
 import math
 from functools import partial
diff --git a/timm/models/swin_transformer.py b/timm/models/swin_transformer.py
index 92057902..584d5649 100644
--- a/timm/models/swin_transformer.py
+++ b/timm/models/swin_transformer.py
@@ -4,6 +4,7 @@ A PyTorch impl of : `Swin Transformer: Hierarchical Vision Transformer using Shi
 
 Code/weights from https://github.com/microsoft/Swin-Transformer, original copyright/license info below
 
+Modifications and additions for timm hacked together by / Copyright 2021, Ross Wightman
 """
 # --------------------------------------------------------
 # Swin Transformer
diff --git a/timm/models/vgg.py b/timm/models/vgg.py
index b2fe07c8..ccaa21da 100644
--- a/timm/models/vgg.py
+++ b/timm/models/vgg.py
@@ -180,7 +180,7 @@ def _filter_fn(state_dict):
 def _create_vgg(variant: str, pretrained: bool, **kwargs: Any) -> VGG:
     cfg = variant.split('_')[0]
     # NOTE: VGG is one of few models with stride==1 features w/ 6 out_indices [0..5]
-    out_indices = kwargs.get('out_indices', (0, 1, 2, 3, 4, 5))
+    out_indices = kwargs.pop('out_indices', (0, 1, 2, 3, 4, 5))
     model = build_model_with_cfg(
         VGG, variant, pretrained,
         default_cfg=default_cfgs[variant],
diff --git a/timm/models/visformer.py b/timm/models/visformer.py
index 3a9be06f..62f77304 100644
--- a/timm/models/visformer.py
+++ b/timm/models/visformer.py
@@ -4,6 +4,7 @@ Paper: Visformer: The Vision-friendly Transformer - https://arxiv.org/abs/2104.1
 
 From original at https://github.com/danczs/Visformer
 
+Modifications and additions for timm hacked together by / Copyright 2021, Ross Wightman
 """
 from copy import deepcopy
 
diff --git a/timm/models/vision_transformer.py b/timm/models/vision_transformer.py
index 6e568abf..fec44334 100644
--- a/timm/models/vision_transformer.py
+++ b/timm/models/vision_transformer.py
@@ -20,7 +20,7 @@ for some einops/einsum fun
 * Simple transformer style inspired by Andrej Karpathy's https://github.com/karpathy/minGPT
 * Bert reference code checks against Huggingface Transformers and Tensorflow Bert
 
-Hacked together by / Copyright 2021 Ross Wightman
+Hacked together by / Copyright 2020, Ross Wightman
 """
 import math
 import logging
diff --git a/timm/models/vision_transformer_hybrid.py b/timm/models/vision_transformer_hybrid.py
index d5f0a537..d46297ef 100644
--- a/timm/models/vision_transformer_hybrid.py
+++ b/timm/models/vision_transformer_hybrid.py
@@ -11,7 +11,7 @@ A PyTorch implement of the Hybrid Vision Transformers as described in:
 NOTE These hybrid model definitions depend on code in vision_transformer.py.
 They were moved here to keep file sizes sane.
 
-Hacked together by / Copyright 2021 Ross Wightman
+Hacked together by / Copyright 2020, Ross Wightman
 """
 from copy import deepcopy
 from functools import partial
diff --git a/timm/models/xcit.py b/timm/models/xcit.py
index ac5e802c..9ad162ed 100644
--- a/timm/models/xcit.py
+++ b/timm/models/xcit.py
@@ -1,10 +1,12 @@
 """ Cross-Covariance Image Transformer (XCiT) in PyTorch
 
-Same as the official implementation, with some minor adaptations.
-    - https://github.com/facebookresearch/xcit/blob/master/xcit.py
-
 Paper:
     - https://arxiv.org/abs/2106.09681
+
+Same as the official implementation, with some minor adaptations, original copyright below
+    - https://github.com/facebookresearch/xcit/blob/master/xcit.py
+
+Modifications and additions for timm hacked together by / Copyright 2021, Ross Wightman
 """
 # Copyright (c) 2015-present, Facebook, Inc.
 # All rights reserved.