An Analysis of Chinese LLM Censorship and Bias with Qwen 2 Instruct
•
48
lhl PRO
leonardlin
AI & ML interests
None yet
Recent Activity
liked
a dataset
about 20 hours ago
openai/webgpt_comparisons
liked
a dataset
about 20 hours ago
Dahoas/synthetic-instruct-gptj-pairwise
updated
a model
21 days ago
kinokokoro/cyberagent-mistral-nemo-webnovels
Articles
Organizations
Posts
7
Post
1851
My weekened project ended up being doing some testing between torchtune, axolotl, and unsloth. I *think* it's a 1:1 comparison of what LoRA fine-tuning performance looks like between the different hardware I have in my dev boxes (4090, 3090, 7900 XTX, W7900) with a few other interesting tidbits.
Tonight I wrote up a WandB report (the panel editor is super broken in Firefox 😔) that sums up some of the more interesting bits from the results: https://wandb.ai/augmxnt/train-bench/reports/torchtune-vs-axolotl-vs-unsloth-Trainer-Comparison--Vmlldzo4MzU3NTAx
Tonight I wrote up a WandB report (the panel editor is super broken in Firefox 😔) that sums up some of the more interesting bits from the results: https://wandb.ai/augmxnt/train-bench/reports/torchtune-vs-axolotl-vs-unsloth-Trainer-Comparison--Vmlldzo4MzU3NTAx
Post
2426
Maybe of interest, I just finished a long writeup of my weekend project exploring Qwen 2 7B Instruct's Chinese censorship: https://huggingface.co/blog/leonardlin/chinese-llm-censorship-analysis
I also have an accompanying model and dataset (and codebase) for those curious to poke around:
* augmxnt/Qwen2-7B-Instruct-deccp
* augmxnt/deccp
I also have an accompanying model and dataset (and codebase) for those curious to poke around:
* augmxnt/Qwen2-7B-Instruct-deccp
* augmxnt/deccp
Collections
24
-
LLM in a flash: Efficient Large Language Model Inference with Limited Memory
Paper • 2312.11514 • Published • 258 -
PowerInfer: Fast Large Language Model Serving with a Consumer-grade GPU
Paper • 2312.12456 • Published • 41 -
Accelerating LLM Inference with Staged Speculative Decoding
Paper • 2308.04623 • Published • 23 -
LLM.int8(): 8-bit Matrix Multiplication for Transformers at Scale
Paper • 2208.07339 • Published • 4
-
QuIP: 2-Bit Quantization of Large Language Models With Guarantees
Paper • 2307.13304 • Published • 2 -
SpQR: A Sparse-Quantized Representation for Near-Lossless LLM Weight Compression
Paper • 2306.03078 • Published • 3 -
OmniQuant: Omnidirectionally Calibrated Quantization for Large Language Models
Paper • 2308.13137 • Published • 17 -
AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration
Paper • 2306.00978 • Published • 8
spaces
1
models
None public yet
datasets
None public yet