© 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. 2-RES-SM
    11. 27414
    Prev
    Next
    INDEX
    Explanations

    structured data related to programming and coding 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/2-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.2.hook_resid_post
    Hook Layer
    2
    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
    450
    +0.14
    0.8%
    508
    +0.13
    0.8%
    24
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    136
    +0.14
    -0.02
    7
    +0.13
    0.04
    60
    +0.12
    0.06
    Negative Logits
    rob
    -1.47
    she
    -1.33
    lish
    -1.30
     worthy
    -1.30
    plots
    -1.30
    http
    -1.29
    >`
    -1.29
    chool
    -1.28
    girl
    -1.24
    orna
    -1.20
    POSITIVE LOGITS
    TRODUCTION
    1.30
    ppings
    1.28
     faith
    1.25
    urname
    1.23
    NON
    1.22
     heirs
    1.21
     discretion
    1.19
     beard
    1.19
     tracts
    1.18
     CURIAM
    1.18
    Activations Density 2.814%

    No Known Activations