© 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. 9146
    Prev
    Next
    INDEX
    Explanations

    mentions of legal agreements and contracts

    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
    271
    +0.15
    0.9%
    78
    +0.12
    0.7%
    397
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    443
    +0.15
    0.01
    59
    +0.12
    0.01
    371
    +0.12
    0.01
    Negative Logits
    atrix
    -2.02
    aries
    -1.57
     arriv
    -1.49
     ways
    -1.41
    rates
    -1.41
    rapeut
    -1.40
    elijk
    -1.34
     vu
    -1.34
     brightest
    -1.34
     chambers
    -1.31
    POSITIVE LOGITS
    ÏħÏĦ
    1.65
    ãĥ¥
    1.55
    ÑĥÑģ
    1.46
    овоÑĢ
    1.44
    license
    1.40
    ολ
    1.38
    .<
    1.37
    Ñĥм
    1.37
    )\].
    1.34
    ãĥĥãĥĪ
    1.34
    Activations Density 0.026%

    No Known Activations