Pinned Repositories
PaLM-pytorch
Implementation of the specific Transformer architecture from PaLM - Scaling Language Modeling with Pathways
FLAN
LaMDA-pytorch
Open-source pre-training implementation of Google's LaMDA in PyTorch.
OGEYRRAT
Open-source Gloriously Extensive Yaml-configuration Repository for Reimplementing Architectures of Transformers
augmenting-conv-data
Just testing
axolotl
Go ahead and axolotl questions
convogpt
Conversational Language model toolkit for training against human preferences.
DeepSeek-MoE
dreamer-for-text
An experiment
json2binidx_tool
TearGosling's Repositories
TearGosling/OWARIDA
Open reWritten, Augmented and (sometimes) Reversed Instruction Data for All
TearGosling/augmenting-conv-data
Just testing
TearGosling/dreamer-for-text
An experiment
TearGosling/whisper.cpp
Port of OpenAI's Whisper model in C/C++
TearGosling/nanoGPT_aru
The simplest, fastest repository for training/finetuning medium-sized GPTs.
TearGosling/DeepSeek-MoE
TearGosling/RWKV-LM
RWKV is an RNN with transformer-level LLM performance. It can be directly trained like a GPT (parallelizable). So it's combining the best of RNN and transformer - great performance, fast inference, saves VRAM, fast training, "infinite" ctx_len, and free sentence embedding.
TearGosling/RWKV-infctx-trainer
RWKV infctx trainer, for training arbitary context sizes, to 10k and beyond!
TearGosling/learnable-merging
TearGosling/LyCORIS
Lora beYond Conventional methods, Other Rank adaptation Implementations for Stable diffusion.
TearGosling/axolotl
Go ahead and axolotl questions
TearGosling/json2binidx_tool
TearGosling/mezo-pretraining-test
TearGosling/MeetInTheMiddle-Finetuning
Meet-in-the-Middle training method, but for fine-tuning
TearGosling/OGEYRRAT
Open-source Gloriously Extensive Yaml-configuration Repository for Reimplementing Architectures of Transformers
TearGosling/U-ViT_JEPA
A PyTorch implementation of the paper "All are Worth Words: A ViT Backbone for Diffusion Models".
TearGosling/FLAN
TearGosling/VNDataConverter
Grabbing parsed VNs of many different formats and converting it into a universal format into a universal
TearGosling/convogpt
Conversational Language model toolkit for training against human preferences.
TearGosling/LaMDA-pytorch
Open-source pre-training implementation of Google's LaMDA in PyTorch.
TearGosling/smoothquant
SHORKIN ALL THE BUGSSSS AWAY
TearGosling/torch-int
AGHHHHH
TearGosling/PaLM-pytorch
Implementation of the specific Transformer architecture from PaLM - Scaling Language Modeling with Pathways