© 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. 10764
    Prev
    Next
    INDEX
    Explanations

    references to numerical limits or maximums in various contexts

    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
    247
    +0.13
    0.7%
    307
    +0.12
    0.7%
    479
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    85
    +0.13
    0.03
    267
    +0.12
    0.03
    13
    +0.12
    0.01
    Negative Logits
    wart
    -2.06
    rer
    -1.90
    rers
    -1.69
    forth
    -1.67
    cker
    -1.59
    cup
    -1.59
    rier
    -1.55
    FROM
    -1.54
    ous
    -1.54
    hors
    -1.54
    POSITIVE LOGITS
     length
    2.24
    imal
    2.24
     diameter
    2.16
     sized
    2.16
     amount
    2.14
    į
    2.12
     size
    2.08
    ime
    2.06
     lifespan
    2.04
     permissible
    1.99
    Activations Density 0.280%

    No Known Activations