© 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. 8874
    Prev
    Next
    INDEX
    Explanations

    occurrences of URLs and links in the text

    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/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
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    21
    +0.34
    2.1%
    78
    +0.14
    0.9%
    478
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    21
    +0.34
    0.02
    395
    +0.14
    0.03
    219
    +0.12
    0.03
    Negative Logits
     measured
    -1.63
    hold
    -1.59
     larger
    -1.59
     toward
    -1.58
     sequenced
    -1.49
     examined
    -1.48
     towards
    -1.48
     healthy
    -1.47
     studied
    -1.47
     seasons
    -1.46
    POSITIVE LOGITS
    ://
    4.25
    :/
    1.82
    fiddle
    1.80
    apis
    1.76
    docs
    1.72
    sdl
    1.71
    jsfiddle
    1.69
    adoc
    1.69
    rq
    1.63
    dbc
    1.61
    Activations Density 0.029%

    No Known Activations