© 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. 4-RES-SM
    11. 9420
    Prev
    Next
    INDEX
    Explanations

    phrases of gratitude and appreciation

    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
    307
    +0.25
    1.5%
    306
    +0.24
    1.4%
    447
    +0.16
    1.0%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    447
    +0.25
    0.04
    485
    +0.24
    0.03
    306
    +0.16
    0.03
    Negative Logits
    ®
    -2.55
    İ
    -2.38
    ĩ
    -2.36
    ¡
    -2.25
    »¿
    -2.24
    Ľ
    -2.22
    <|outofrange|>
    -2.19
    ↵
    -2.19
    -2.19
    -2.19
    POSITIVE LOGITS
     advice
    2.02
     your
    1.94
     clarification
    1.86
     sharing
    1.85
     info
    1.68
     posting
    1.68
     clar
    1.65
     feedback
    1.63
     downloading
    1.61
     pointing
    1.57
    Activations Density 0.208%

    No Known Activations