© 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. 24854
    Prev
    Next
    INDEX
    Explanations

    instances of the word "refer"

    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
    478
    +0.14
    0.8%
    63
    +0.14
    0.8%
    185
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    63
    +0.14
    0.01
    484
    +0.14
    0.01
    472
    +0.12
    0.01
    Negative Logits
    «
    -2.88
    ½
    -2.77
    Ħ
    -2.75
    º
    -2.71
    Ľ
    -2.60
    ĥ½
    -2.56
    ²
    -2.51
    µ
    -2.39
    ¤
    -2.38
    Ĥ¬
    -2.37
    POSITIVE LOGITS
    è¨Ģ
    1.51
    ably
    1.48
     primarily
    1.47
     me
    1.47
    acional
    1.47
     apologies
    1.46
     surname
    1.43
     mystical
    1.38
    abel
    1.33
     coffin
    1.31
    Activations Density 0.029%

    No Known Activations