© 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. 1-RES-SM
    11. 28865
    Prev
    Next
    INDEX
    Explanations

    verbs related to exploration and discovery

    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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    156
    +0.19
    1.1%
    23
    +0.19
    1.1%
    376
    +0.17
    1.0%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    111
    +0.19
    0.07
    156
    +0.19
    0.05
    148
    +0.17
    0.04
    Negative Logits
    rael
    -1.88
    ball
    -1.68
    enberg
    -1.56
    arium
    -1.54
     asleep
    -1.53
    ία
    -1.52
    imes
    -1.52
    eps
    -1.52
    inki
    -1.51
    ension
    -1.48
    POSITIVE LOGITS
     how
    1.59
     them
    1.59
    them
    1.56
    Ļª
    1.54
    ³
    1.51
    ¢
    1.49
     crowds
    1.48
    sects
    1.46
     the
    1.39
    ľĵ
    1.37
    Activations Density 0.419%

    No Known Activations