Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Feat: GPT2 Inference Infrastructure (GGUF, StateDict, BPE Tokenizer) #237

Merged
merged 29 commits into from
Nov 17, 2024
Merged

Feat: GPT2 Inference Infrastructure (GGUF, StateDict, BPE Tokenizer) #237

merged 29 commits into from
Nov 17, 2024

Conversation

hikettei
Copy link
Owner

@hikettei hikettei commented Nov 16, 2024 •
edited
Loading

  • caten/apps
  • Restore Parameter from GGUF/PyTorch
  • CLI Tool (Roswell)
  • [ x download the model from https://huggingface.co/{model_size}/resolve/main/pytorch_model.bin
  • Upload GGUF+GPT2 to huggingface, restore the weight from GGUF
  • Need to implement load_state_dict
  • adding load-weight method for Module/Class?
  • gguf status
    • gpt2
    • gpt2-medium
    • gpt2-large
    • gpt2-xl
    • make public
    • load from url
  • (next pr) update the impl of gpt2
  • Tokenizer: GGUF -> Caten BPE
  • Restore the weight from GGUF
  • with-eval not to initialize the weight for defmodule slots

@hikettei hikettei changed the title Feat: Implement GPT2 Feat: GPT2 Inference Infrastructure Nov 17, 2024
@hikettei
Copy link
Owner Author

more easier way to avoid full symbolic is to add kv cache and fix s=1?

@hikettei hikettei mentioned this pull request Nov 16, 2024
Closed
28 tasks
@hikettei hikettei changed the title Feat: GPT2 Inference Infrastructure Feat: GPT2 Inference Infrastructure (GGUF, StateDict, BPE) Nov 17, 2024
@hikettei hikettei changed the title Feat: GPT2 Inference Infrastructure (GGUF, StateDict, BPE) Feat: GPT2 Inference Infrastructure (GGUF, StateDict, BPE Tokenizer) Nov 17, 2024
@hikettei hikettei marked this pull request as ready for review November 17, 2024 12:05
@hikettei hikettei merged commit a608867 into main Nov 17, 2024
6 checks passed
@hikettei hikettei deleted the applications branch November 17, 2024 12:26
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
1 participant
@hikettei