© 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. 15112
    Prev
    Next
    INDEX
    Explanations

    specific software and version references related to development environments

    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/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
    Neuron Alignment
    Index
    Value
    % of L₁
    263
    +0.13
    0.7%
    239
    +0.13
    0.7%
    321
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    146
    +0.13
    0.04
    98
    +0.13
    0.01
    239
    +0.13
    0.03
    Negative Logits
    ĵ
    -4.86
    ħ
    -4.78
    ·¸
    -4.61
    ĥ½
    -4.56
    ´
    -4.56
    ij
    -4.52
    ·
    -4.50
    Ļª
    -4.50
    IJ
    -4.50
    ¬
    -4.47
    POSITIVE LOGITS
     Version
    2.22
     IDE
    1.99
     Code
    1.96
     Developer
    1.89
    gradle
    1.88
     Console
    1.84
     Solution
    1.84
     Office
    1.83
     2013
    1.82
     2008
    1.80
    Activations Density 0.186%

    No Known Activations