Hugging Face
Models
Datasets
Spaces
Posts
Docs
Enterprise
Pricing
Log In
Sign Up
13
3
lisass123
linsa11
Follow
lzZzZx328's profile picture
1 follower
ยท
3 following
AI & ML interests
None yet
Recent Activity
reacted
to
etemiz
's
post
with ๐ค
19 days ago
-= DeepSeek V3 =- After installing the new CUDA toolkit and compiling llama.cpp again I tested DeepSeek V3 yesterday. In terms of human alignment DeepSeek V3 did worse on: - health - fasting - nostr - misinfo - nutrition did better on: - faith - bitcoin - alternative medicine - ancient wisdom compared to DeepSeek 2.5. In my opinion overall it is worse than 2.5. And 2.5 wasn't that great. There is a general tendency of models getting smarter but at the same time getting less wiser, less human aligned, less beneficial to humans. I don't know what is causing this. But maybe synthetic dataset use for further training the LLMs makes it more and more detached from humanity. This is not going in the right direction. My solution is to come up with a curator council to determine the datasets that are closest to human preference. "Humans that care about other humans the most" could be a definition of this dataset. What do you think?
reacted
to
etemiz
's
post
with ๐
19 days ago
-= DeepSeek V3 =- After installing the new CUDA toolkit and compiling llama.cpp again I tested DeepSeek V3 yesterday. In terms of human alignment DeepSeek V3 did worse on: - health - fasting - nostr - misinfo - nutrition did better on: - faith - bitcoin - alternative medicine - ancient wisdom compared to DeepSeek 2.5. In my opinion overall it is worse than 2.5. And 2.5 wasn't that great. There is a general tendency of models getting smarter but at the same time getting less wiser, less human aligned, less beneficial to humans. I don't know what is causing this. But maybe synthetic dataset use for further training the LLMs makes it more and more detached from humanity. This is not going in the right direction. My solution is to come up with a curator council to determine the datasets that are closest to human preference. "Humans that care about other humans the most" could be a definition of this dataset. What do you think?
upvoted
an
article
19 days ago
Mastering Tensor Dimensions in Transformers
View all activity
Organizations
None yet
linsa11
's activity
All
Models
Datasets
Spaces
Papers
Collections
Community
Posts
Upvotes
Likes
liked
3 models
4 months ago
nvidia/NVLM-D-72B
Image-Text-to-Text
โข
Updated
18 days ago
โข
46.8k
โข
766
stepfun-ai/GOT-OCR2_0
Image-Text-to-Text
โข
Updated
Sep 18, 2024
โข
613k
โข
1.35k
nvidia/OpenMath2-Llama3.1-8B
Text Generation
โข
Updated
Nov 25, 2024
โข
1.35k
โข
26