© 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. 32481
    Prev
    Next
    INDEX
    Explanations

    specific roles and positions related to leadership and service, particularly in artistic and governmental contexts

    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
    198
    +0.14
    0.8%
    218
    +0.14
    0.8%
    500
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    198
    +0.14
    0.07
    218
    +0.14
    0.07
    249
    +0.13
    0.03
    Negative Logits
    emia
    -1.74
     <
    -1.64
    iga
    -1.57
    gmail
    -1.50
    uria
    -1.40
    mmas
    -1.39
    >\
    -1.39
    ENRON
    -1.37
    jection
    -1.35
     BGCOLOR
    -1.34
    POSITIVE LOGITS
     protector
    1.61
     anchor
    1.48
     indicator
    1.44
     attach
    1.41
    ariate
    1.38
    hero
    1.37
     affairs
    1.36
     grows
    1.36
     absor
    1.35
    unto
    1.34
    Activations Density 1.252%

    No Known Activations