© 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. 32029
    Prev
    Next
    INDEX
    Explanations

    structured data inputs and their formatting

    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
    125
    +0.14
    0.8%
    478
    +0.14
    0.8%
    274
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    177
    +0.14
    0.02
    133
    +0.14
    0.02
    230
    +0.13
    0.02
    Negative Logits
    ing
    -1.88
    zes
    -1.75
    area
    -1.70
    ressing
    -1.66
    istry
    -1.64
    iquity
    -1.53
    ding
    -1.50
    ausing
    -1.48
    ed
    -1.46
    stick
    -1.45
    POSITIVE LOGITS
    Īĺ
    3.23
    ĻĤ
    3.23
    į
    3.15
    ı
    2.94
    ĨĴ
    2.93
    İ
    2.88
    ĩ
    2.88
    ®
    2.73
    ħ
    2.65
    ĭ
    2.64
    Activations Density 0.074%

    No Known Activations