© 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. 30847
    Prev
    Next
    INDEX
    Explanations

    terms related to directional movement and orientation

    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
    160
    +0.13
    0.8%
    296
    +0.12
    0.7%
    104
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    433
    +0.13
    0.01
    104
    +0.12
    0.01
    116
    +0.12
    0.01
    Negative Logits
    sburg
    -1.78
     faith
    -1.56
    pshire
    -1.53
    ilent
    -1.52
    ourt
    -1.52
     agon
    -1.48
     court
    -1.47
     fuss
    -1.46
     pleasure
    -1.45
     ransom
    -1.42
    POSITIVE LOGITS
    ality
    2.56
    ally
    2.53
    mapsto
    2.04
    ī
    2.01
     direction
    1.99
     perpendicular
    1.95
    ably
    1.94
    imetry
    1.85
    IJ
    1.83
    edly
    1.82
    Activations Density 0.115%

    No Known Activations