© 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. 2-RES-SM
    11. 10420
    Prev
    Next
    INDEX
    Explanations

    sentences that describe opportunities for registration or joining programs

    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
    316
    +0.12
    0.8%
    72
    +0.11
    0.7%
    278
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    259
    +0.12
    0.03
    316
    +0.11
    0.05
    414
    +0.10
    0.05
    Negative Logits
     published
    -1.42
     corresponding
    -1.36
     partition
    -1.35
     participating
    -1.34
     reasons
    -1.34
     newly
    -1.34
     binding
    -1.33
     underwent
    -1.32
     respective
    -1.30
     voluntarily
    -1.28
    POSITIVE LOGITS
    noop
    1.57
    osph
    1.49
    ypse
    1.44
    ytics
    1.44
    zzo
    1.40
    issance
    1.38
    uru
    1.38
    ина
    1.37
    oka
    1.36
    sein
    1.35
    Activations Density 5.989%

    No Known Activations