© 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. 4-RES-SM
    11. 8678
    Prev
    Next
    INDEX
    Explanations

    terms related to media and entertainment entities

    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/4-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.4.hook_resid_post
    Hook Layer
    4
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Neuron Alignment
    Index
    Value
    % of L₁
    156
    +0.64
    4.5%
    111
    +0.15
    1.1%
    137
    +0.09
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    419
    +0.64
    0.08
    416
    +0.15
    0.07
    205
    +0.09
    0.07
    Negative Logits
    ĥ½
    -2.26
    ¬
    -1.81
    Ĵ
    -1.72
    IJ
    -1.69
    ¸
    -1.62
    ¥
    -1.61
    ģ
    -1.55
    Ĭ
    -1.52
    ½
    -1.51
    -1.51
    POSITIVE LOGITS
    rosse
    1.44
    Inflater
    1.44
    agne
    1.41
    elines
    1.36
    ists
    1.33
    rics
    1.33
    ist
    1.33
    etine
    1.32
    neb
    1.32
    agin
    1.30
    Activations Density 0.496%

    No Known Activations