© 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. 10147
    Prev
    Next
    INDEX
    Explanations

    prepositions indicating direction or connection

    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
    332
    +0.13
    0.7%
    483
    +0.12
    0.6%
    69
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    483
    +0.13
    0.03
    427
    +0.12
    0.03
    371
    +0.11
    0.03
    Negative Logits
    Ĵ
    -3.30
    Ļª
    -3.11
    ¢
    -3.00
    ·¸
    -2.95
    Ľ
    -2.92
    ģ
    -2.91
    ¡
    -2.75
    Ł
    -2.71
    ¨
    -2.62
    ij
    -2.60
    POSITIVE LOGITS
     adulthood
    2.42
     2010
    2.38
     2016
    2.34
     2009
    2.33
     2017
    2.30
     2015
    2.26
     2006
    2.24
     2008
    2.23
     2012
    2.23
     2013
    2.21
    Activations Density 0.108%

    No Known Activations