Fetching metadata from the HF Docker repository...
-
1.61 kB
update for hugging face
-
64 Bytes
add openwebtext based gpt model
-
1.07 kB
make streamlit app
-
6.85 kB
test
-
11.4 kB
add basic BiGram testing
-
5.49 kB
add basic BiGram testing
-
2.25 kB
update extract script
-
13.6 kB
simple gpt model based on entire works of shakespeare
-
8.01 kB
simple gpt model based on entire works of shakespeare
model.pkl
Detected Pickle imports (16)
- "__main__.MultiHeadAttention",
- "collections.OrderedDict",
- "__main__.GPTLanguageModel",
- "torch.nn.modules.container.ModuleList",
- "torch.nn.modules.normalization.LayerNorm",
- "torch.nn.modules.activation.ReLU",
- "torch._utils._rebuild_tensor_v2",
- "torch.nn.modules.linear.Linear",
- "torch.nn.modules.dropout.Dropout",
- "__main__.Block",
- "torch.nn.modules.container.Sequential",
- "__main__.Head",
- "torch.storage._load_from_bytes",
- "torch._utils._rebuild_parameter",
- "torch.nn.modules.sparse.Embedding",
- "__main__.FeedFoward"
How to fix it?
160 MB
add model pickle file
-
58 Bytes
update for hugging face
-
47 Bytes
update for hugging face
-
5.62 MB
add shakespeare text and rename main script to testing script
-
8.68 kB
update train_gpt_openwebtext script
-
133 kB
add vocab to repo