© 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. 9445
    Prev
    Next
    INDEX
    Explanations

    references to private entities or organizations

    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
    438
    +0.14
    0.8%
    365
    +0.14
    0.8%
    454
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    438
    +0.14
    0.02
    208
    +0.14
    0.03
    247
    +0.13
    0.02
    Negative Logits
    IJ
    -2.75
    ½
    -2.59
    »
    -2.58
    ģ
    -2.56
    Ī
    -2.54
    ·¸
    -2.44
    ª
    -2.41
    İ
    -2.41
    Ģ
    -2.41
    ħ
    -2.38
    POSITIVE LOGITS
     sector
    2.39
     equity
    2.08
     affairs
    1.91
     property
    1.89
     residence
    1.82
    sector
    1.76
     quarters
    1.74
     estate
    1.74
     practice
    1.74
     Sector
    1.71
    Activations Density 0.214%

    No Known Activations