© 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. 1-RES-SM
    11. 29422
    Prev
    Next
    INDEX
    Explanations

    mentions of pediatric topics and related medical specialties

    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/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
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    156
    +0.24
    1.4%
    87
    +0.14
    0.8%
    111
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    13
    +0.24
    0.01
    87
    +0.14
    0.01
    404
    +0.11
    0.01
    Negative Logits
     direct
    -1.58
    uben
    -1.53
    ayed
    -1.49
     true
    -1.47
     trusting
    -1.43
     ''
    -1.42
     realised
    -1.42
     redeem
    -1.40
     refined
    -1.40
     ‘
    -1.40
    POSITIVE LOGITS
    º
    2.84
    ·¸
    2.70
    ĸ
    2.61
    §
    2.58
    Ĩ
    2.53
    Į
    2.52
    £
    2.51
    ĸ´
    2.49
    ł
    2.44
    Ĥ
    2.43
    Activations Density 0.049%

    No Known Activations