© 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. 4-RES-SM
    11. 12475
    Prev
    Next
    INDEX
    Explanations

    web addresses or URLs, particularly those ending in ".com"

    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
    397
    +0.22
    1.3%
    466
    +0.17
    1.0%
    307
    +0.16
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    500
    +0.22
    0.03
    353
    +0.17
    0.03
    425
    +0.16
    0.04
    Negative Logits
    oretic
    -1.39
    ¯
    -1.36
     plague
    -1.35
    documentclass
    -1.35
    ĩ
    -1.31
     balls
    -1.31
    holds
    -1.30
    than
    -1.30
     diseases
    -1.28
    Ĭ
    -1.26
    POSITIVE LOGITS
    /~
    3.48
    /?
    3.08
    /
    2.93
    /_
    2.84
    /
    2.76
    /@
    2.72
    /#
    2.60
    /.
    2.53
    /,
    2.53
    /)
    2.39
    Activations Density 0.072%

    No Known Activations