Vojtech Cekal
cekal
AI & ML interests
In love with machine learning.
Organizations
cekal's activity
Different answer after each request
2
#130 opened 7 months ago
by
amin2557
Mistral 7B produces different results when we hit via postman api
7
#124 opened 7 months ago
by
DivyaKanniah
Discussion Thread - Let's Get To Know the (Allegedly) Leaked "mistral-medium" better!
2
#4 opened 7 months ago
by
cekal
![](https://cdn-avatars.huggingface.co/v1/production/uploads/63ebfd71ca08a72ba9ce6fe0/WEXOVko_Lgvq_Y8_Zlb4o.png)
How to fine-tune this? + Training code
43
#19 opened 9 months ago
by
cekal
![](https://cdn-avatars.huggingface.co/v1/production/uploads/63ebfd71ca08a72ba9ce6fe0/WEXOVko_Lgvq_Y8_Zlb4o.png)
Responses by phi-2 are off, it's depressed and insults me for no reason whatsover?
7
#61 opened 8 months ago
by
NightcoreSkies
How to use system prompt?
1
#69 opened 8 months ago
by
mznw
Is there any simple way to solve the problem of redundant output
3
#68 opened 8 months ago
by
jjplane
How many GPUs do we need to run this out of box?
3
#63 opened 8 months ago
by
kz919
![](https://cdn-avatars.huggingface.co/v1/production/uploads/62140dcdcf7928035e8135ad/FTiirwS_L6IaLHmHwIo2g.png)
WARNING:root:Some parameters are on the meta device device because they were offloaded to the cpu.
4
#25 opened 9 months ago
by
kmukeshreddy
![](https://cdn-avatars.huggingface.co/v1/production/uploads/1665881381855-noauth.jpeg)
FINE TUNING with PEFT MIXTRAL
2
#17 opened 9 months ago
by
TK4000
PEFT based Fine Tuned model hallucinates values from the fine tuning training data while inferencing.
7
#111 opened 9 months ago
by
Pradeep1995
![](https://cdn-avatars.huggingface.co/v1/production/uploads/1599822346546-noauth.jpeg)
Delete model-00010-of-00019.safetensors
1
#21 opened 9 months ago
by
dynamicmortal
Freaking outstanding! GPT-4 confirms
2
#1 opened 9 months ago
by
TioPeperino
Support for LoRA?
17
#3 opened over 1 year ago
by
cekal
![](https://cdn-avatars.huggingface.co/v1/production/uploads/63ebfd71ca08a72ba9ce6fe0/WEXOVko_Lgvq_Y8_Zlb4o.png)
Which dataset was used to train this model?
4
#1 opened over 1 year ago
by
cekal
![](https://cdn-avatars.huggingface.co/v1/production/uploads/63ebfd71ca08a72ba9ce6fe0/WEXOVko_Lgvq_Y8_Zlb4o.png)
Hallucinates too much
4
#6 opened over 1 year ago
by
WajihUllahBaig
The model is a troll >D
1
#17 opened over 1 year ago
by
pplanel
Update modeling_mpt.py
5
#11 opened over 1 year ago
by
huseinzol05
![](https://cdn-avatars.huggingface.co/v1/production/uploads/1646710842120-5e73316106936008a9ee6523.jpeg)
query_key_value for LoRA? (target modules)
2
#10 opened over 1 year ago
by
cekal
![](https://cdn-avatars.huggingface.co/v1/production/uploads/63ebfd71ca08a72ba9ce6fe0/WEXOVko_Lgvq_Y8_Zlb4o.png)
๐ฉ Report โ Ethical Issue: Questions about Discriminative Licensing & Unfair Business Practice
11
#5 opened over 1 year ago
by
alexjc
![](https://cdn-avatars.huggingface.co/v1/production/uploads/1663942087766-63234b1970edc53f51c9a672.jpeg)