© 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. 26066
    Prev
    Next
    INDEX
    Explanations

    references to geographical or political borders

    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
    228
    +0.14
    0.8%
    243
    +0.13
    0.7%
    93
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    228
    +0.14
    0.02
    93
    +0.13
    0.02
    243
    +0.12
    0.02
    Negative Logits
     hereby
    -1.64
     slightest
    -1.59
     tons
    -1.51
     thee
    -1.50
     naught
    -1.50
     days
    -1.47
     little
    -1.45
     pleasure
    -1.45
    ±
    -1.44
     helpless
    -1.41
    POSITIVE LOGITS
    rup
    2.19
    lands
    1.96
    lock
    1.93
    loader
    1.90
    lace
    1.88
    witz
    1.79
    strip
    1.77
    ography
    1.76
    azole
    1.75
    ström
    1.73
    Activations Density 0.108%

    No Known Activations