© 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. 32413
    Prev
    Next
    INDEX
    Explanations

    technical specifications and parameters related to software systems

    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
    263
    +0.20
    1.1%
    271
    +0.16
    0.9%
    326
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    136
    +0.20
    -0.07
    8
    +0.16
    0.05
    263
    +0.14
    0.14
    Negative Logits
     hemat
    -1.40
    [/
    -1.39
     ÂĹ
    -1.35
     *[
    -1.31
     ***
    -1.29
     Memorandum
    -1.27
     materially
    -1.26
     posters
    -1.26
     »
    -1.25
     tents
    -1.25
    POSITIVE LOGITS
    cil
    1.57
    Generated
    1.47
    agment
    1.43
    aring
    1.42
    icable
    1.41
    ().
    1.38
    Derived
    1.37
    weak
    1.35
    EventListener
    1.34
    ENTIAL
    1.33
    Activations Density 4.175%

    No Known Activations