Motoki Wu

Serialization of Attention Layer

1

This might be a Keras problem but have you tried serializing some of the layers? I tried the following to save a model that contains `SoftAttention`: ``` from keras.engine import...

Is the Dependency (child sum tree) option working?

19

Hi, thanks for the great port. I tried `python sentiment.py` after getting the data. When I set `DEPENDENCY == True` I get an assertion error. `````` tree_rnn $ python sentiment.py...

[BUG] Pythia (GPT-NeoX based) models degrade in generation quality using DeepSpeed Inference

2

**Describe the bug** Hi, the GPT-NeoX based [Pythia](https://github.com/EleutherAI/pythia) model's generation quality is degraded once optimized with DeepSpeed. Edit (2023/03/07): related to https://github.com/microsoft/DeepSpeed/issues/2777. **To Reproduce** ```python from transformers import pipeline import...

Bug: `fatal: not in a git directory`

12

## What happened Ran the `actions-ecosystem/action-push-tag@v1` Github Action unsuccessfully: ``` ... "/var/run/docker.sock":"/var/run/docker.sock" -v "/home/runner/work/_temp/_github_home":"/github/home" -v "/home/runner/work/_temp/_github_workflow":"/github/workflow" -v "/home/runner/work/_temp/_runner_file_commands":"/github/file_commands" -v "/home/runner/work/.../...":"/github/workspace" ... 12 fatal: not in a git directory ``` Probably...

**Describe the bug** Running inference on Llama 3 models gives an error ```shell /src/chat-ai/venv3.10/lib/python3.10/site-packages/deepspeed/module_inject/replace_module.py in _replace_module(model, policies, prefix, layer_id, level_id, state_dict) 687 for name, child in model.named_children(): 688 if child.__class__...