© 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. 4-RES-SM
    11. 14232
    Prev
    Next
    INDEX
    Explanations

    references to scientific studies or findings

    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
    306
    +0.12
    0.7%
    266
    +0.11
    0.6%
    398
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    266
    +0.12
    0.03
    258
    +0.11
    0.03
    471
    +0.11
    0.03
    Negative Logits
    arum
    -1.72
    posts
    -1.62
    uuid
    -1.55
    esian
    -1.53
    hers
    -1.48
    prising
    -1.44
    illion
    -1.44
    cology
    -1.43
    manif
    -1.43
    ...](
    -1.40
    POSITIVE LOGITS
    Âł
    1.95
     Berger
    1.88
     Martinez
    1.85
    nez
    1.81
     Kim
    1.79
     Abe
    1.77
     Müller
    1.75
     Schmidt
    1.75
    ¬
    1.74
     Garcia
    1.74
    Activations Density 0.077%

    No Known Activations