canopylabs/orpheus-3b-0.1-ft
Text-to-Speech
β’
Updated
β’
91.4k
β’
β’
542
No Name
Ainonake
AI & ML interests
None yet
Recent Activity
new activity
3 days ago
Undi95/QwQ-RP:I notice need Qwen2, but for SillyTavern, what I need to use?
new activity
4 days ago
TheDrummer/Llama-3SOME-8B-v2:fp8
new activity
7 days ago
Undi95/MLewd-ReMM-L2-Chat-20B:How to query for multi chat
Organizations
None yet
Ainonake's activity
I notice need Qwen2, but for SillyTavern, what I need to use?
3
#2 opened 4 days ago
by
RGTails
How to query for multi chat
2
#5 opened 8 days ago
by
Drseltsam
Ollama deployment
2
1
#7 opened 14 days ago
by
sedatkaradag
Ud quants pleaseπ₯Ί
2
#5 opened 13 days ago
by
Ainonake
Q8 is noticeably better than Q4_0. Both suffer from repetition in rp in non-english lanuages
1
#2 opened 13 days ago
by
Ainonake
LMstudio error
5
#1 opened 13 days ago
by
DmitryV
I am very interested in this dataset
3
138
#3 opened 26 days ago
by
Demanin
Gave it a whirl
1
#4 opened 19 days ago
by
SkyStach
Multiple language support
1
2
#14 opened 19 days ago
by
thunder-007
Report: Nice job, you've ruined a nice model with censorship no one wants or asks for
9
16
#7 opened 21 days ago
by
Ainonake
Can someone explain why this model is currently at top 3 of the trending list?
9
10
#8 opened 20 days ago
by
RobinFeng123
π© Report: Illegal or restricted content
6
10
#1 opened about 1 month ago
by
clecho52
liked
a
model
about 1 month ago
liked
a
model
about 2 months ago
reacted to
tomaarsen's
post with β€οΈ
2 months ago
Post
6697
An assembly of 18 European companies, labs, and universities have banded together to launch πͺπΊ EuroBERT! It's a state-of-the-art multilingual encoder for 15 European languages, designed to be finetuned for retrieval, classification, etc.
πͺπΊ 15 Languages: English, French, German, Spanish, Chinese, Italian, Russian, Polish, Portuguese, Japanese, Vietnamese, Dutch, Arabic, Turkish, Hindi
3οΈβ£ 3 model sizes: 210M, 610M, and 2.1B parameters - very very useful sizes in my opinion
β‘οΈ Sequence length of 8192 tokens! Nice to see these higher sequence lengths for encoders becoming more common.
βοΈ Architecture based on Llama, but with bi-directional (non-causal) attention to turn it into an encoder. Flash Attention 2 is supported.
π₯ A new Pareto frontier (stronger *and* smaller) for multilingual encoder models
π Evaluated against mDeBERTa, mGTE, XLM-RoBERTa for Retrieval, Classification, and Regression (after finetuning for each task separately): EuroBERT punches way above its weight.
π Detailed paper with all details, incl. data: FineWeb for English and CulturaX for multilingual data, The Stack v2 and Proof-Pile-2 for code.
Check out the release blogpost here: https://huggingface.co/blog/EuroBERT/release
* EuroBERT/EuroBERT-210m
* EuroBERT/EuroBERT-610m
* EuroBERT/EuroBERT-2.1B
The next step is for researchers to build upon the 3 EuroBERT base models and publish strong retrieval, zero-shot classification, etc. models for all to use. I'm very much looking forward to it!
πͺπΊ 15 Languages: English, French, German, Spanish, Chinese, Italian, Russian, Polish, Portuguese, Japanese, Vietnamese, Dutch, Arabic, Turkish, Hindi
3οΈβ£ 3 model sizes: 210M, 610M, and 2.1B parameters - very very useful sizes in my opinion
β‘οΈ Sequence length of 8192 tokens! Nice to see these higher sequence lengths for encoders becoming more common.
βοΈ Architecture based on Llama, but with bi-directional (non-causal) attention to turn it into an encoder. Flash Attention 2 is supported.
π₯ A new Pareto frontier (stronger *and* smaller) for multilingual encoder models
π Evaluated against mDeBERTa, mGTE, XLM-RoBERTa for Retrieval, Classification, and Regression (after finetuning for each task separately): EuroBERT punches way above its weight.
π Detailed paper with all details, incl. data: FineWeb for English and CulturaX for multilingual data, The Stack v2 and Proof-Pile-2 for code.
Check out the release blogpost here: https://huggingface.co/blog/EuroBERT/release
* EuroBERT/EuroBERT-210m
* EuroBERT/EuroBERT-610m
* EuroBERT/EuroBERT-2.1B
The next step is for researchers to build upon the 3 EuroBERT base models and publish strong retrieval, zero-shot classification, etc. models for all to use. I'm very much looking forward to it!
- 1 reply
This shit is fire
13
#2 opened 2 months ago
by
Ainonake
liked
a
model
2 months ago
Then what if we do the same, but put whole conversation in first user input?
So it will be
System prompt
User: conversation history
Then ask R1 to generate thinking.
And the amount of messages in conversation history should be varied. Then, Bot reply will always contain thinking.
What do you think about doing part of the dataset with replies from some context?
E.g. we have e.g. 50% of data with thinking from first user answer, and some parts of dataset with
User,
Bot (no thinking),
User
Bot (no thinking),
User, N times,
Then ask R1 to think here and train on it. So the model will understand long context better.