datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
qwen-base-error-analysis
Qwen3.5-4B-Base Error Analysis Dataset
This dataset contains 10 diverse examples where the base language model Qwen3.5-4B-Base makes incorrect or unexpected predictions. It was created as part of an exploration of base model blind spots and failure modes.
Model Tested
Model: Qwen/Qwen3.5-4B-Base
Type: Pre-trained base model (not instruction-tuned)
Parameters: 4B
Architecture: Causal Language Model with hybrid Gated DeltaNet + Attention layers
Dataset Structure… See the full description on the dataset page: https://huggingface.co/datasets/ArslanMZahid/qwen-base-error-analysis.Qwen3.5-0.8B-Base-error-analysis
Qwen3.5-0.8B-Base Error Analysis Dataset
Model Tested
Model: Qwen/Qwen3.5-0.8B-Base
Model Type: Base / Pretrained Causal Language Model
Parameters: 0.8B
Architecture: Causal Language Model with Vision Encoder (Gated Delta Networks + MoE)
Release Date: February 2026
License: Apache 2.0
Dataset Description
This dataset contains 10 input prompts where Qwen/Qwen3.5-0.8B-Base produced incorrect, incomplete, or hallucinated outputs. The prompts were designed… See the full description on the dataset page: https://huggingface.co/datasets/AhsanShahid/Qwen3.5-0.8B-Base-error-analysis.