Fine-Tuning xLSTM-7B on a budget: An Experimental Journey with Chat Templates
• 1
ethicalabs.ai
non-profit
Verified
AI & ML interests
Building practical, ethical and sustainable AI/ML projects.
Recent Activity
View all activity
Articles
ethicalabs.ai's entry in the OpenAIRE AI Hackathon 2026, co-organised by OpenAIRE and Alien Intelligence — a 12-week open science build challenge
-
ethicalabs/Echo-DSRN-v0.1.3-Research-Intent-CLF
Text Classification • 98.3M • Updated • 579 -
Echo DSRN SmolTools 114M NSFW Classifier
👀Echo-DSRN 114M SmolTools: NSFW Intent Classifier (English)
-
ethicalabs/Echo-SmolTools-114M-NSFW-CLF
Text Classification • 98.3M • Updated • 76 -
ethicalabs/Echo-SmolTools-114M-NSFW-CLF-PEFT
Text Generation • Updated • 40
Qwen2.5-0.5B-Instruct (backbone) + Echo-DSRN (memory)
The Empathetic Assistant.
-
ethicalabs/Kurtis-E1.1-Qwen3-4B
Text Generation • 4B • Updated • 25 -
ethicalabs/Kurtis-E1.1-Qwen3-4B-GGUF
Text Generation • 4B • Updated • 32 • 1 -
ethicalabs/Kurtis-E1.1-Qwen2.5-3B-Instruct
Text Generation • 3B • Updated • 17 • • 1 -
ethicalabs/Kurtis-E1.1-Qwen2.5-0.5B-Instruct
Text Generation • 0.5B • Updated • 9 •
A Multimodal Deep Learning Approach for Skin Cancer Classification by using ViTs (Visual Transformers)
-
ethicalabs/Flwr-Qwen3-4B-Coding-PEFT
Text Generation • Updated • 3 -
ethicalabs/FlowerTune-Qwen2.5-Coder-0.5B-Instruct-PEFT
Text Generation • Updated • 9 -
ethicalabs/Flwr-Phi-4-mini-Instruct-Coding-PEFT
Text Generation • Updated • 3 • 1 -
ethicalabs/Flwr-SmolLM2-1.7B-Instruct-Coding-PEFT
Text Generation • Updated • 1
Granite 4.0 Language Models for macOS (MLX-LM) - 4-bit Quantization
Dual-State Recurrent Neural Network
-
ethicalabs/Echo-DSRN-114M-v0.1.2
Text Generation • 0.1B • Updated • 509 -
ethicalabs/Echo-DSRN-114M-v0.1.2-Base
Text Generation • 0.1B • Updated • 16 -
Echo DSRN Small Next Word Prediction
🔥Echo-DSRN-114M (114M RNN LM) - Next Word Prediction
-
Echo-DSRN-114M Semantic Compressor Demo (CPU)
🗜2Echo-DSRN-114M Semantic Compressor Demo (CPU)
Transformer backbone + Echo-DSRN (RNN)
The Empathetic Assistant (v1.2)
Synthetic Datasets for SLMs fine-tuning.
Self-Improving LLMs Through Iterative Refinement
ethicalabs.ai's entry in the OpenAIRE AI Hackathon 2026, co-organised by OpenAIRE and Alien Intelligence — a 12-week open science build challenge
-
ethicalabs/Echo-DSRN-v0.1.3-Research-Intent-CLF
Text Classification • 98.3M • Updated • 579 -
Echo DSRN SmolTools 114M NSFW Classifier
👀Echo-DSRN 114M SmolTools: NSFW Intent Classifier (English)
-
ethicalabs/Echo-SmolTools-114M-NSFW-CLF
Text Classification • 98.3M • Updated • 76 -
ethicalabs/Echo-SmolTools-114M-NSFW-CLF-PEFT
Text Generation • Updated • 40
Dual-State Recurrent Neural Network
-
ethicalabs/Echo-DSRN-114M-v0.1.2
Text Generation • 0.1B • Updated • 509 -
ethicalabs/Echo-DSRN-114M-v0.1.2-Base
Text Generation • 0.1B • Updated • 16 -
Echo DSRN Small Next Word Prediction
🔥Echo-DSRN-114M (114M RNN LM) - Next Word Prediction
-
Echo-DSRN-114M Semantic Compressor Demo (CPU)
🗜2Echo-DSRN-114M Semantic Compressor Demo (CPU)
Qwen2.5-0.5B-Instruct (backbone) + Echo-DSRN (memory)
Transformer backbone + Echo-DSRN (RNN)
The Empathetic Assistant.
-
ethicalabs/Kurtis-E1.1-Qwen3-4B
Text Generation • 4B • Updated • 25 -
ethicalabs/Kurtis-E1.1-Qwen3-4B-GGUF
Text Generation • 4B • Updated • 32 • 1 -
ethicalabs/Kurtis-E1.1-Qwen2.5-3B-Instruct
Text Generation • 3B • Updated • 17 • • 1 -
ethicalabs/Kurtis-E1.1-Qwen2.5-0.5B-Instruct
Text Generation • 0.5B • Updated • 9 •
The Empathetic Assistant (v1.2)
A Multimodal Deep Learning Approach for Skin Cancer Classification by using ViTs (Visual Transformers)
Synthetic Datasets for SLMs fine-tuning.
-
ethicalabs/Flwr-Qwen3-4B-Coding-PEFT
Text Generation • Updated • 3 -
ethicalabs/FlowerTune-Qwen2.5-Coder-0.5B-Instruct-PEFT
Text Generation • Updated • 9 -
ethicalabs/Flwr-Phi-4-mini-Instruct-Coding-PEFT
Text Generation • Updated • 3 • 1 -
ethicalabs/Flwr-SmolLM2-1.7B-Instruct-Coding-PEFT
Text Generation • Updated • 1
Granite 4.0 Language Models for macOS (MLX-LM) - 4-bit Quantization
Self-Improving LLMs Through Iterative Refinement