© 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. 26024
    Prev
    Next
    INDEX
    Explanations

    references to a specific person or term associated with the name "Han."

    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
    410
    +0.17
    0.9%
    180
    +0.11
    0.6%
    269
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    410
    +0.17
    0.04
    191
    +0.11
    0.03
    43
    +0.10
    0.03
    Negative Logits
    orters
    -1.73
     little
    -1.61
    unctional
    -1.43
     discriminate
    -1.39
    ubation
    -1.39
    atement
    -1.36
     testis
    -1.34
    EOF
    -1.34
     late
    -1.33
     hereinafter
    -1.32
    POSITIVE LOGITS
    ess
    2.15
    jan
    1.93
    ji
    1.88
    çois
    1.85
    agar
    1.81
    onymous
    1.80
    amycin
    1.80
    iewicz
    1.79
    estamp
    1.77
    esses
    1.76
    Activations Density 0.194%

    No Known Activations