diff options
| author | Daniel Bevenius <daniel.bevenius@gmail.com> | 2024-02-13 14:15:42 +0100 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2024-02-13 15:15:42 +0200 |
| commit | 263978904c7472db1865409a7ff1129599f6a40b (patch) | |
| tree | 9c6f6f7732f474c74a9a1eafe8b52bcd8936d221 /gguf-py | |
| parent | cf45252a7cfcb998bade46a886e20477cecc538a (diff) | |
finetune : rename feed-forward tensors (w1/w2/w3) (#4839)
* finetune: rename feed-forward tensors (w1/w2/w3)
This commit renames the feed-forward tensors w1, w2 and w3 to ffn_gate,
ffn_down and ffn_up respectively.
The motivation for this change is to make it easier to understand the
purpose of the tensors. This also seems to be inline with the names
used in the llama_layer struct in llama.cpp.
Signed-off-by: Daniel Bevenius <daniel.bevenius@gmail.com>
* train-text-from-scratch: rename ff tensors
This commit renames the feed-forward tensors w1, w2 and w3 to ffn_gate,
ffn_down and ffn_up respectively.
The motivation for this change is to make it easier to understand the
purpose of the tensors. This also seems to be inline with the names
used in the llama_layer struct in llama.cpp
Signed-off-by: Daniel Bevenius <daniel.bevenius@gmail.com>
---------
Signed-off-by: Daniel Bevenius <daniel.bevenius@gmail.com>
Diffstat (limited to 'gguf-py')
0 files changed, 0 insertions, 0 deletions