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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 26522
    Prev
    Next
    INDEX
    Explanations

    references to locks, unlocking, and related mechanisms

    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
    156
    +0.19
    1.1%
    260
    +0.18
    1.1%
    173
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    260
    +0.19
    0.03
    304
    +0.18
    0.03
    173
    +0.14
    0.02
    Negative Logits
    ¹
    -2.11
    ↵↵
    -1.95
    -1.95
    ↵
    -1.95
    -1.95
    ↵
    -1.95
    <|outofrange|>
    -1.95
    -1.95
    <|outofrange|>
    -1.95
    -1.95
    POSITIVE LOGITS
    hart
    2.50
    ingly
    2.36
    heed
    2.25
    down
    2.14
    estone
    1.96
    mith
    1.92
    ership
    1.85
     horns
    1.80
    ateral
    1.75
     doors
    1.72
    Activations Density 0.165%

    No Known Activations