Hongye Jin

EarthWorm001's activity

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

New activity in inclusionAI/Ling-mini-base-2.0 13 days ago

Is the 20T ckpt annealed? Mid-trained? Do you mind provide more details?

#1 opened 13 days ago by

EarthWorm001

New activity in google/gemma-7b-it about 2 years ago

gemma-7b-it doesn't answer for some questions and returns '/n'

11

#55 opened about 2 years ago by

mudogruer

New activity in google/gemma-7b about 2 years ago

Weird completions

3

#39 opened about 2 years ago by

rodrigo-nogueira

Weird answer

4

#40 opened about 2 years ago by

KunAndKun

New activity in google/gemma-7b-it about 2 years ago

<pad> spam issue

13

#40 opened about 2 years ago by

Zewsic

New activity in google/gemma-7b about 2 years ago

Very different results with float16. [Actually, gemma-7b-it does not work with float16]

👍 3

6

#33 opened about 2 years ago by

EarthWorm001

New activity in mistralai/Mistral-7B-v0.1 over 2 years ago

Is SWA used during pertaining?

🤝 2

#113 opened over 2 years ago by

EarthWorm001