-
alessiodevoto/exp_att_stats_h2oai_h2o-danube3-500m-chat_kmfoda_booksum_100_1000_4
2.38M • Updated • 838 -
alessiodevoto/exp_att_stats_Qwen_Qwen3-4B-Instruct-2507_kmfoda_booksum_100_1000_4
19M • Updated • 827 -
alessiodevoto/exp_att_stats_meta-llama_Llama-3.2-1B-Instruct_kmfoda_booksum_100_1000_4
16.9M • Updated • 800 -
alessiodevoto/exp_att_stats_MaxJeblick_llama2-0b-unit-test_kmfoda_booksum_100_1000_4
168 • Updated • 23.1k
Alessio
alessiodevoto
AI & ML interests
Efficient training/inference - Adaptive computation - Interpretability
Recent Activity
new activity 5 days ago
alessiodevoto/math500:Add boxed final answer instruction 🤖🤖🤖 new activity 5 days ago
alessiodevoto/aime25:Add boxed final answer instruction 🤖🤖🤖 upvoted a paper 10 months ago
Attention Sinks in Diffusion Language ModelsOrganizations
-
alessiodevoto/exp_att_stats_Qwen_Qwen3-8B_kmfoda_booksum_100_4000_4
19M • Updated • 3 -
alessiodevoto/exp_att_stats_Qwen_Qwen3-8B_kmfoda_booksum_100_100_4
19M • Updated • 3 -
alessiodevoto/exp_att_stats_Qwen_Qwen3-8B_kmfoda_booksum_10_500_4
19M • Updated • 5 -
alessiodevoto/exp_att_stats_google_gemma-2-9b-it_kmfoda_booksum_100_4000_4
44.2M • Updated • 3
-
alessiodevoto/exp_att_stats_h2oai_h2o-danube3-500m-chat_kmfoda_booksum_100_1000_4
2.38M • Updated • 838 -
alessiodevoto/exp_att_stats_Qwen_Qwen3-4B-Instruct-2507_kmfoda_booksum_100_1000_4
19M • Updated • 827 -
alessiodevoto/exp_att_stats_meta-llama_Llama-3.2-1B-Instruct_kmfoda_booksum_100_1000_4
16.9M • Updated • 800 -
alessiodevoto/exp_att_stats_MaxJeblick_llama2-0b-unit-test_kmfoda_booksum_100_1000_4
168 • Updated • 23.1k
-
alessiodevoto/exp_att_stats_Qwen_Qwen3-8B_kmfoda_booksum_100_4000_4
19M • Updated • 3 -
alessiodevoto/exp_att_stats_Qwen_Qwen3-8B_kmfoda_booksum_100_100_4
19M • Updated • 3 -
alessiodevoto/exp_att_stats_Qwen_Qwen3-8B_kmfoda_booksum_10_500_4
19M • Updated • 5 -
alessiodevoto/exp_att_stats_google_gemma-2-9b-it_kmfoda_booksum_100_4000_4
44.2M • Updated • 3