© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 32462
    Prev
    Next
    INDEX
    Explanations

    references to 'position' in various contexts

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Neuron Alignment
    Index
    Value
    % of L₁
    498
    +0.17
    0.9%
    404
    +0.12
    0.7%
    502
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    498
    +0.17
    0.03
    46
    +0.12
    0.01
    426
    +0.12
    0.02
    Negative Logits
    ª
    -3.56
    ĵ
    -3.51
    ¼
    -3.36
    ł
    -3.30
    Ĵ
    -3.27
    ķ
    -3.26
    ľĵ
    -3.22
    ĻĤ
    -3.20
    ½
    -3.08
    ¿
    -3.08
    POSITIVE LOGITS
     operator
    1.63
    ally
    1.60
    picker
    1.59
    imetry
    1.59
    unov
    1.52
    ova
    1.52
     vacated
    1.50
    inator
    1.49
    ater
    1.49
    ellar
    1.48
    Activations Density 0.140%

    No Known Activations