© 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. 1-RES-SM
    11. 29468
    Prev
    Next
    INDEX
    Explanations

    punctuation marks indicating significant closures or separations in 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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    258
    +0.14
    0.8%
    369
    +0.14
    0.8%
    203
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    441
    +0.14
    0.02
    418
    +0.14
    0.02
    202
    +0.13
    0.03
    Negative Logits
    iNdEx
    -1.77
    xspace
    -1.66
    otide
    -1.53
    oks
    -1.49
    "></
    -1.47
    zzles
    -1.46
     echo
    -1.44
    imon
    -1.44
    igner
    -1.43
    Usage
    -1.41
    POSITIVE LOGITS
    sworth
    1.61
     sides
    1.45
     appearance
    1.43
    she
    1.40
    imir
    1.39
     wonders
    1.38
    sen
    1.37
     Hill
    1.34
     supp
    1.31
     outdoors
    1.31
    Activations Density 0.277%

    No Known Activations