ByteDance
/

Ouro-1.4B

Text Generation

looped-language-model

recurrent-depth

Model card Files Files and versions

Resources

View closed (2)

Fix bos/eos token IDs (config.json + tokenizer_config.json)

#11 opened 7 days ago by

Fix UniversalTransformerCache.get_mask_sizes for batched generation

#10 opened 7 days ago by

Batched generation (batch_size > 1) produces incorrect outputs — possible causal mask issue?

#9 opened about 2 months ago by

Differences in the results of the reproduction test on lm-evaluation-harness

#8 opened 2 months ago by

difference definition of eos token id in config.json and tokenizer_config.json

#7 opened 3 months ago by

Inference with ChatML Template

#6 opened 3 months ago by

LoRA Merge - Package Available

#4 opened 3 months ago by

Fix for KV Cache Bug - ouro-cache-fix Package

#3 opened 4 months ago by

Bytedance_ouro-1.4B.ipynb Colab Notebook

#2 opened 4 months ago by