© 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. 12686
    Prev
    Next
    INDEX
    Explanations

    terms related to operating systems and their functionality

    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
    110
    +0.14
    0.8%
    172
    +0.14
    0.8%
    470
    +0.14
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    110
    +0.14
    0.03
    304
    +0.14
    0.02
    445
    +0.14
    0.02
    Negative Logits
    xton
    -1.62
    wort
    -1.60
    ounder
    -1.59
    jamin
    -1.55
    dimen
    -1.55
    noreply
    -1.51
    ples
    -1.50
    iday
    -1.48
    usalem
    -1.48
    weed
    -1.48
    POSITIVE LOGITS
     theat
    2.20
     smoothly
    2.10
     independently
    2.07
    ively
    2.05
     rooms
    1.97
     remotely
    1.91
     theatre
    1.90
     theaters
    1.89
     machinery
    1.88
     hours
    1.86
    Activations Density 0.274%

    No Known Activations