© 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 EvalsExportsAPI 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. 11212
    Prev
    Next
    INDEX
    Explanations

    instances of the word "to."

    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
    447
    +0.17
    1.0%
    411
    +0.14
    0.8%
    430
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    411
    +0.17
    0.03
    434
    +0.14
    0.03
    447
    +0.14
    0.02
    Negative Logits
    »¿
    -3.87
    ĻĤ
    -3.64
    Ļª
    -3.05
    Ļ
    -2.98
    ĸ´
    -2.97
    Ķ
    -2.96
    ľĵ
    -2.91
    ¾
    -2.87
    Ĥ¬
    -2.85
    ½
    -2.81
    POSITIVE LOGITS
     proceed
    2.27
     exit
    1.81
     quit
    1.81
     begin
    1.79
     participate
    1.74
     register
    1.71
     join
    1.70
     resume
    1.69
     bypass
    1.68
     enter
    1.59
    Activations Density 0.092%

    No Known Activations