Amgad Hasan
amgadhasan
AI & ML interests
LLM
ASR
Recent Activity
updated
a model
7 days ago
amgadhasan/phi-4
upvoted
an
article
20 days ago
Power steering: Squeeze massive power from small LLMs
Organizations
amgadhasan's activity
Which infra to run openai/whisper-large-v3-turbo ?
1
#7 opened 3 months ago
by
Jack0o
how many GPU memory do I need to finetune largeV3
4
#150 opened 4 months ago
by
lanejohn
What is the accuracy of the model? Why can't I fine-tune when I set the accuracy to FP16?
1
#157 opened 4 months ago
by
chengligen
How to fine tuning with timestamps
2
#122 opened 4 months ago
by
deepdml
Max sequence length
3
#2 opened 4 months ago
by
amgadhasan
llama3.1 version
1
#3 opened 5 months ago
by
amgadhasan
Delete original
#1 opened 5 months ago
by
amgadhasan
What is the difference between this and official meta repo?
3
#9 opened 5 months ago
by
fzyzcjy
Issues with Fine-Tuning
9
#21 opened 5 months ago
by
ironrock
Phi-3.1?
1
#1 opened 5 months ago
by
amgadhasan
The model stops after generating one new token
2
#76 opened 6 months ago
by
rajiv-data-chef
Why the lower temperature?
2
#4 opened 6 months ago
by
amgadhasan
Wrong chat template
5
#1 opened 6 months ago
by
amgadhasan
Prompting Style?
1
#3 opened 6 months ago
by
xriminact
Add support for chat tempalte in llama.cpp
2
#1 opened 6 months ago
by
amgadhasan
Where is v1?
#2 opened 6 months ago
by
amgadhasan
Prompt Template?
3
#2 opened 6 months ago
by
amgadhasan
Add info about the prompt template
1
#4 opened 6 months ago
by
amgadhasan
Add chat template from mistralai/Mistral-7B-Instruct-v0.2
#3 opened 6 months ago
by
amgadhasan
Typo in winrates figure
1
#2 opened 7 months ago
by
amgadhasan