© 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. 2-RES-SM
    11. 27880
    Prev
    Next
    INDEX
    Explanations

    syntax elements in a programming or markup language

    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
    478
    +0.15
    0.8%
    0
    +0.13
    0.7%
    138
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    242
    +0.15
    0.03
    0
    +0.13
    0.03
    124
    +0.12
    0.03
    Negative Logits
    .’”
    -1.92
    .[]{
    -1.88
    amycin
    -1.83
    *]{}.
    -1.71
    oplasma
    -1.62
    obacter
    -1.61
    []{
    -1.60
    ablo
    -1.58
    .’
    -1.57
    .]{}
    -1.56
    POSITIVE LOGITS
    ¿
    4.83
    ·
    4.76
    ĵ
    4.75
    ¾
    4.57
    Īĺ
    4.55
    Ĥ¬
    4.54
    ¥
    4.51
    ©
    4.51
    §
    4.50
    ¿½
    4.50
    Activations Density 0.103%

    No Known Activations