Massimo Roberto Scamarcia PRO
mrs83
AI & ML interests
Natural Language Processing, Text Generation, Question Answering, Data Augmentation, Knowledge Transfer, Chain-of-Thought, ResearchOps, MLOps
Recent Activity
updated a model about 18 hours ago
ethicalabs/Echo-DSRN-114M updated a model about 18 hours ago
ethicalabs/Echo-DSRN-114M-v0.1.1 published a model about 22 hours ago
ethicalabs/Echo-DSRN-114M-v0.1.1Organizations
🚨 BREAKING: Echo-DSRN-114M downloads Pizza via HTTP
#1 opened 3 days ago
by 
mrs83
Training stability on AMD ROCm (distributed backend, gloo vs nccl)
#3 opened 4 days ago
by mrs83
Continued Pre-Training (SFTTrainer, HuggingFaceH4/ultrachat_200k)
#2 opened 6 days ago
by mrs83
Truncated Backpropagation Through Time (TBPTT) on Fineweb-EDU (10BT)
#1 opened 7 days ago
by mrs83
Echo-DSRN - Triton Kernel Benchmark Report - PyTorch (native) vs Triton Legacy (sequential) vs Triton 3-Pass (new)
1
#10 opened 29 days ago
by mrs83
SFT/Alignment - Phase 007-06-MLP8: ethicalabs/Kurtis-EON1-SFT Mix (1 epoch)
2
#9 opened 29 days ago
by mrs83
SFT/Alignment - Phase 007-02-MLP8: mlabonne/FineTome-100k-dedup (1000 steps + 500 steps)
1
#8 opened about 1 month ago
by mrs83
SFT/Alignment - Phase 007-02: mlabonne/FineTome-100k-dedup (Filtered, 128T, 1 Epoch)
#7 opened about 1 month ago
by mrs83
SFT/Alignment - Phase 007: Smoltalk2 (No Thinking, Daily Conversations)
1
#6 opened about 1 month ago
by mrs83
Mid-Training - Phase 006: Smoltalk2 (No Thinking)
4
#5 opened about 2 months ago
by mrs83
Mid-Training - Phase 005: Kurtis SFT Mix + OpenHermes2.5
5
#3 opened about 2 months ago
by mrs83
A quick update on the development of Kurtis-EON1 (Echo-DSRN)
1
#4 opened about 2 months ago
by mrs83
Mid-Training - Phase 004: Kurtis SFT Mix
9
#2 opened about 2 months ago
by mrs83
Mid-Training - Phase 003: HuggingFaceTB/smoltalk
4
#1 opened about 2 months ago
by mrs83
Hunyuan-Mamba hybrid out 2502 - Where is the Mistral-xLSTM hybrid (DeepSeek moment)?
17
#11 opened about 1 year ago
by 
gue22
Hunyuan-Mamba hybrid out 2502 - Where is the Mistral-xLSTM hybrid (DeepSeek moment)?
17
#11 opened about 1 year ago
by gue22
tensorblock/ethicalabs_Kurtis-E1.1-Qwen2.5-3B-Instruct-GGUF
#5499 opened 6 months ago
by mrs83
LLM tasks routing
#2 opened 8 months ago
by mrs83
Diversity of responses
👍🚀 1
2
#2 opened 8 months ago
by 
entfane