Kristian Schwethelm

KristianS7's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

New activity in KristianS7/prepacked-fineweb-edu-llama2-32K-T2048 4 days ago

Add paper, project page and code links

#1 opened 4 days ago by

nielsr

New activity in KristianS7/Ouro-2.6B 3 months ago

Transformers v5 compatibility

#1 opened 3 months ago by

KristianS7

New activity in KristianS7/Ouro-1.4B 5 months ago

Add assistant generation tags to chat template

#4 opened 5 months ago by

KristianS7

Update tied weight metadata for Transformers 5

#3 opened 5 months ago by

KristianS7

Update Ouro remote code for Transformers 5.9

1

#2 opened 5 months ago by

KristianS7

Fix default RoPE initialization for Transformers 5.9

#1 opened 5 months ago by

KristianS7

New activity in ByteDance/Ouro-2.6B 8 months ago

Lower evaluation results

1

#2 opened 11 months ago by

MianchuWang

New activity in ByteDance/Ouro-1.4B 8 months ago

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

3

#8 opened 10 months ago by

ThreeGold116

New activity in ByteDance/Ouro-2.6B 8 months ago

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

#5 opened 8 months ago by

KristianS7

Fix UniversalTransformerCache.get_mask_sizes for batched generation

#4 opened 8 months ago by

KristianS7

New activity in ByteDance/Ouro-1.4B 8 months ago

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

#11 opened 8 months ago by

KristianS7

Fix UniversalTransformerCache.get_mask_sizes for batched generation

👍 1

#10 opened 8 months ago by

KristianS7

New activity in ByteDance/Ouro-1.4B-Thinking 9 months ago

Fix UniversalTransformerCache.get_mask_sizes for batched generation

1

#5 opened 9 months ago by

KristianS7

New activity in ByteDance/Ouro-2.6B-Thinking 9 months ago

Fix UniversalTransformerCache.get_mask_sizes for batched generation

1

#8 opened 9 months ago by

KristianS7

New activity in ByteDance/Ouro-1.4B 9 months ago

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

➕ 1

1

#9 opened 9 months ago by

vconchel

New activity in ByteDance/Ouro-2.6B-Thinking 9 months ago

Fix bos/eos token IDs + add enable_thinking to chat template

2

#7 opened 9 months ago by

KristianS7

New activity in ByteDance/Ouro-1.4B-Thinking 9 months ago

Fix bos/eos token IDs + add enable_thinking to chat template

2

#4 opened 9 months ago by

KristianS7

New activity in ByteDance/Ouro-2.6B-Thinking 9 months ago

Fix bos/eos token IDs + add enable_thinking to chat template

2

#7 opened 9 months ago by

KristianS7

New activity in ByteDance/Ouro-1.4B-Thinking 9 months ago

Fix bos/eos token IDs + add enable_thinking to chat template

2

#4 opened 9 months ago by

KristianS7