SUMIN/HuggingFace_transformer
1
0
Fork 0
Code Issues Pull Requests Actions 7 Packages Projects Releases Wiki Activity

Update all references to canonical models (#29001)

Browse Source 
* Script & Manual edition

* Update
This commit is contained in:
Lysandre Debut
2024-02-16 08:16:58 +01:00
committed by GitHub
parent 1e402b957d
commit f497f564bb
561 changed files with 2682 additions and 2687 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
docs/source/en/model_doc/gpt_bigcode.md
View File

@@ -38,7 +38,7 @@ The main differences compared to GPT2.
- Use jit to fuse the attention fp32 casting, masking, softmax, and scaling.
- Combine the attention and causal masks into a single one, pre-computed for the whole model instead of every layer.
- Merge the key and value caches into one (this changes the format of layer_past/ present, does it risk creating problems?)
- Use the memory layout (self.num_heads, 3, self.head_dim) instead of `(3, self.num_heads, self.head_dim)` for the QKV tensor with MHA. (prevents an overhead with the merged key and values, but makes the checkpoints incompatible with the original gpt2 model).
- Use the memory layout (self.num_heads, 3, self.head_dim) instead of `(3, self.num_heads, self.head_dim)` for the QKV tensor with MHA. (prevents an overhead with the merged key and values, but makes the checkpoints incompatible with the original openai-community/gpt2 model).
You can read more about the optimizations in the [original pull request](https://github.com/huggingface/transformers/pull/22575)