© 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. 14157
    Prev
    Next
    INDEX
    Explanations

    punctuation marks

    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
    303
    +0.13
    0.7%
    478
    +0.13
    0.7%
    23
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    23
    +0.13
    0.02
    271
    +0.13
    0.00
    136
    +0.12
    -0.00
    Negative Logits
    hops
    -1.68
    ozy
    -1.68
    amble
    -1.58
    ier
    -1.48
    ign
    -1.47
    resp
    -1.47
    ixel
    -1.43
    erialize
    -1.42
     parity
    -1.39
    Xiv
    -1.38
    POSITIVE LOGITS
    IJ
    2.11
    ĻĤ
    1.75
     grandson
    1.74
     '')
    1.66
     Esq
    1.62
    himself
    1.61
    ĨĴ
    1.61
     son
    1.60
     his
    1.60
     nephew
    1.60
    Activations Density 0.040%

    No Known Activations