© 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. 1-RES-SM
    11. 27576
    Prev
    Next
    INDEX
    Explanations

    the repetition of the word "already" as a marker of prior completion or existence

    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
    335
    +0.12
    0.7%
    376
    +0.11
    0.6%
    413
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    111
    +0.12
    0.09
    156
    +0.11
    0.05
    365
    +0.10
    0.05
    Negative Logits
    ĨĴ
    -4.64
    Ļª
    -4.43
    Ĵ
    -4.31
    ©
    -4.09
    §
    -3.92
    ĸ´
    -3.71
    ¡
    -3.69
    Ģ
    -3.67
    ³
    -3.64
    į
    -3.64
    POSITIVE LOGITS
     spent
    1.66
    here
    1.62
     PUBL
    1.61
     numbered
    1.57
     married
    1.55
    far
    1.55
     labelled
    1.55
     occupied
    1.55
    ?’
    1.52
     gone
    1.52
    Activations Density 0.543%

    No Known Activations