abidlabs/mnist-loss-spike-demo-trackio-bucket
279 kB
Abubakar Abid PRO
abidlabs
AI & ML interests
self-supervised learning, applications to medicine & biology, interpretation, reproducibility
Recent Activity
updated a model about 2 hours ago
abidlabs/mnist-cnn-loss-spike-demo updated a bucket about 2 hours ago
abidlabs/mnist-loss-spike-demo-trackio-bucket updated a Space about 2 hours ago
abidlabs/mnist-loss-spike-demo-trackioOrganizations
Posts 21
Post
1365
Uhh did Opus 4.8 cheat on PostTrainBench??
it found an API key in the PostTrainBench environment that allowed it to generate synthetic training data without using GPU hours, boosting the base model by 0.4913
Source: https://posttrainbench.com/traces/run.html?id=claude_non_api_max_claude-opus-4-8_10h_run1__healthbench_Qwen_Qwen3-4B-Base_17315102#tab=trace
it found an API key in the PostTrainBench environment that allowed it to generate synthetic training data without using GPU hours, boosting the base model by 0.4913
Source: https://posttrainbench.com/traces/run.html?id=claude_non_api_max_claude-opus-4-8_10h_run1__healthbench_Qwen_Qwen3-4B-Base_17315102#tab=trace
spaces 938
pinned
Runtime error
Agents
Testing Leaderboard
🥇
Duplicate this leaderboard to initialize your own!
pinned
Running on Zero
Agents
1
MiniMax Music 3 Studio
🎵
Generate full songs from structured captions and lyrics
pinned
Running on Zero
Agents
4
MiniMax H3
🎬
Video generation with a synchronized soundtrack
pinned
Running on Zero
Agents
343
Demucs Music Source Separation (v4)
⚡
Separate music into vocals and instrumental
pinned
Running on Zero
Agents
1
MiniCPM-V 4.6 Demo
🪐
MiniCPM-V 4.6 Ultra-Efficient Multimodal AI
pinned
Runtime error
MCP
7
FLUX.1 Kontext
⚡
Kontext image editing on FLUX[dev]
models 38
abidlabs/mnist-cnn-loss-spike-demo
Updated
abidlabs/mnist-cnn-scratch-1min
Updated
abidlabs/mnist-mlp
Image Classification • Updated
abidlabs/mnist-1min
Updated
abidlabs/mnist-from-scratch-1min
Image Classification • Updated
abidlabs/mnist-mlp-1min
Updated
abidlabs/jev-typed-decisions-causal-0.6b
Updated • 57
abidlabs/jev-typed-decisions-causal-0.6b-smoke
Updated
abidlabs/mnist-cnn-test
Image Classification • Updated • 26
abidlabs/gemma-4-E2B-it-distill-sft
Updated
datasets 77
abidlabs/lm-memorization-repro-results
Updated • 123
abidlabs/smelt-repro-materials
Updated • 74
abidlabs/ORCA-sample
Viewer • Updated • 100 • 709 • 1
abidlabs/repro-how-much-can-language-models-memorize-traces
Traces • Updated • 1 • 81
abidlabs/repro-memory-savings-at-what-cost-a-study-of-alternatives-to-backpropagation-traces
Traces • Updated • 1 • 71
abidlabs/repro-memorize-results
Updated • 183
abidlabs/testing-logbook-v2-traces
Traces • Updated • 2 • 63
abidlabs/repro-olaf-world-spectral-assets
Viewer • Updated • 1 • 21
abidlabs/olaf-world-cot-repro
Updated • 51
abidlabs/olaf-world-backprop-repro
Updated • 9