kashifr

5 ms·

1.
Tokenizers v1(huggingface-tokenizers-v1.static.hf.space)
12 points by kashifr 13d ago|0 comments
2.
by kashifr 2mo ago
Check out my pytorch reproduction of the paper here for those interested: https://github.com/NVIDIA/physicsnemo/pull/1660
3.
Carbon: Autoregressive Genomic Foundation Model(huggingface.co)
7 points by kashifr 5mo ago|1 comments
4.
The ultimate guide to RL environments: building and scaling them in the LLM era(huggingface.co)
7 points by kashifr 5mo ago|0 comments
5.
Distilling 100B+ Models 40x Faster with TRL(huggingface.co)
13 points by kashifr 6mo ago|0 comments
6.
by kashifr 6mo ago
You can try out TinyLoRA in PEFT main now: https://huggingface.co/docs/peft/main/en/package_reference/t...
7.
Keep the Tokens Flowing: Lessons from 16 Open-Source RL Libraries(huggingface.co)
2 points by kashifr 7mo ago|0 comments
8.
Transformers V5 is out!(github.com)
10 points by kashifr 8mo ago|0 comments
9.
The Smol Training Playbook: The Secrets to Building World-Class LLMs(huggingface.co)
265 points by kashifr 11mo ago|19 comments
10.
Unlocking On-Policy Distillation for Any Model Family(huggingface.co)
6 points by kashifr 11mo ago|1 comments
11.
Transformers 4.55 New OpenAI GPT OSS(github.com)
2 points by kashifr 1y ago|1 comments
12.
Smollm3: Smol, multilingual, long-context reasoner LLM(huggingface.co)
388 points by kashifr 1y ago|79 comments
13.
by kashifr 1y ago
Have you tried https://geobase.app/ they recently also had a post about duckdb integration: https://geobase.app/blog/duckdb-1-1-3
14.
Epic vs. Apple(twitter.com)
7 points by kashifr 1y ago|0 comments
15.
AIMO (AI Math Olympiad) progress prize winning solution(huggingface.co)
9 points by kashifr 2y ago|0 comments
16.
MaPO: A reference-free alignment technique for diffusion models(mapo-t2i.github.io)
2 points by kashifr 2y ago|1 comments
17.
by kashifr 2y ago
A novel and memory-friendly preference alignment method for diffusion models that does not depend on any reference model.
18.
OpenHermesPreferences: Dataset of ~1M AI preferences from teknium/OpenHermes-2.5(huggingface.co)
7 points by kashifr 3y ago|1 comments
19.
by kashifr 3y ago
The dataset can be used for training preference models or aligning language models through techniques like Direct Preference Optimization (DPO).
20.
HuggingFace Training Cluster as a Service(huggingface.co)
101 points by kashifr 3y ago|45 comments
21.
HuggingFace 235M series D at a $4.5B valuation(twitter.com)
3 points by kashifr 3y ago|0 comments
22.
Fine-tune Llama 2 with DPO(huggingface.co)
3 points by kashifr 3y ago|0 comments
23.
by kashifr 3y ago
An efficient finetuning approach that reduces memory usage enough to finetune a 65B parameter model on a single 48GB GPU while preserving full 16-bit finetuning task performance!
24.
QLoRA 4-bit finetuning of LLMs(github.com)
7 points by kashifr 3y ago|1 comments
25.
by kashifr 3y ago
The title is confusing but refers to the Huggingface Diffusers v0.15 release which brings new pipelines for video and audio to diffusers, showing that diffusion is a great choice for all sorts of generative tasks.
26.
StackLlama: A hands-on guide to train LlaMa with RLHF(huggingface.co)
165 points by kashifr 3y ago|38 comments
27.
by kashifr 3y ago
All the steps involved in training a LlaMa model to answer questions on Stack Exchange data with RLHF.
28.
HuggingFace Diffusers 0.2 with Stable Diffusion pipeline(github.com)
2 points by kashifr 4y ago|1 comments
29.
by kashifr 4y ago
Diffusers now supports CompVis, Stability AI and LAION's Stable Diffusion model weights.
30.
Diffusers: Modular Diffusion model library from HuggingFace(github.com)
47 points by kashifr 4y ago|5 comments

More ›