© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 2-RES-SM
    11. 10931
    Prev
    Next
    INDEX
    Explanations

    references to supreme courts or appellate courts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__res-sm_processed/2-res-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.2.hook_resid_post
    Hook Layer
    2
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    271
    +0.12
    0.7%
    364
    +0.12
    0.6%
    74
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    364
    +0.12
    0.02
    254
    +0.12
    0.02
    104
    +0.11
    0.01
    Negative Logits
    mouth
    -1.61
    oshop
    -1.52
    ypse
    -1.50
    til
    -1.50
    /)
    -1.49
    '">
    -1.44
    figure
    -1.43
    saw
    -1.42
     saturation
    -1.41
    )?
    -1.40
    POSITIVE LOGITS
    ¦
    2.13
     Division
    1.78
    ¯
    1.75
    Ī
    1.75
     Commissioner
    1.72
    ¤
    1.70
    ĸ
    1.65
    ĻĤ
    1.63
    ģ
    1.63
    ·
    1.62
    Activations Density 0.077%

    No Known Activations