© 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. 14424
    Prev
    Next
    INDEX
    Explanations

    numerical data and related values in the text

    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
    +0.17
    1.0%
    146
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    261
    +0.25
    0.03
    306
    +0.17
    0.03
    53
    +0.13
    0.03
    Negative Logits
    ettes
    -1.73
     dominance
    -1.54
     spirits
    -1.41
    GFR
    -1.38
     cohes
    -1.34
     corpus
    -1.32
     bases
    -1.31
     nerves
    -1.31
     DOI
    -1.31
     early
    -1.30
    POSITIVE LOGITS
    ĻĤ
    2.36
    ¦
    2.12
    ·¸
    2.07
    ¢
    1.95
    ĥ½
    1.95
    Ķ
    1.93
    Ī
    1.84
    ĸ´
    1.83
    ·
    1.82
    However
    1.81
    Activations Density 0.137%

    No Known Activations