© 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. 1-RES-SM
    11. 25990
    Prev
    Next
    INDEX
    Explanations

    references to vehicles

    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
    1.2%
    150
    +0.14
    0.9%
    361
    +0.11
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    150
    +0.19
    0.03
    257
    +0.14
    0.02
    194
    +0.11
    0.02
    Negative Logits
    ĥ½
    -2.22
    £
    -1.95
     hope
    -1.84
    ĥ
    -1.73
    Ĺ
    -1.71
    lying
    -1.66
    ·¸
    -1.61
     fellow
    -1.60
    ederal
    -1.52
     theolog
    -1.51
    POSITIVE LOGITS
    ousel
    2.45
    illon
    2.30
    osity
    1.85
    pool
    1.72
    nel
    1.69
    quel
    1.69
    isme
    1.69
    door
    1.60
    ics
    1.58
     comprising
    1.58
    Activations Density 0.106%

    No Known Activations