© 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. 1-RES-SM
    11. 23832
    Prev
    Next
    INDEX
    Explanations

    references to various types of explosive devices and incidents related to violence

    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/1-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.1.hook_resid_post
    Hook Layer
    1
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Neuron Alignment
    Index
    Value
    % of L₁
    156
    +0.26
    1.5%
    166
    +0.12
    0.7%
    454
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    71
    +0.26
    0.04
    156
    +0.12
    0.04
    186
    +0.11
    0.01
    Negative Logits
    ely
    -1.61
    umbent
    -1.61
     grac
    -1.55
    denly
    -1.50
     hers
    -1.48
     kindly
    -1.48
    ably
    -1.43
    inde
    -1.42
    ĥ½
    -1.40
    innati
    -1.39
    POSITIVE LOGITS
     deton
    1.97
     explosions
    1.55
    abad
    1.52
     terrorists
    1.47
    icolor
    1.46
    ameters
    1.43
    scape
    1.42
    inj
    1.38
    ")]
    1.37
     sights
    1.37
    Activations Density 0.276%

    No Known Activations