© 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. 14681
    Prev
    Next
    INDEX
    Explanations

    proper nouns, particularly names of individuals and organizations

    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
    21
    +0.26
    1.7%
    9
    +0.21
    1.4%
    78
    +0.17
    1.1%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    21
    +0.26
    0.12
    59
    +0.21
    0.08
    352
    +0.17
    0.06
    Negative Logits
     respect
    -1.51
     favor
    -1.51
     chance
    -1.47
     OTHERWISE
    -1.44
     outcome
    -1.43
    soever
    -1.43
     course
    -1.43
     likelihood
    -1.43
     selecting
    -1.37
     component
    -1.33
    POSITIVE LOGITS
    ¨
    2.43
    ½
    2.28
    ²
    2.28
    ¹
    2.27
    ħ
    2.24
    ¤
    2.20
    ³
    2.15
    «
    2.14
    Ļ
    2.12
    ª
    2.11
    Activations Density 0.671%

    No Known Activations