© 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. 25659
    Prev
    Next
    INDEX
    Explanations

    terms related to copyright and intellectual property

    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%
    252
    +0.14
    0.8%
    70
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    252
    +0.19
    0.02
    298
    +0.14
    0.02
    486
    +0.12
    0.02
    Negative Logits
    ·
    -1.60
    ²
    -1.58
     departing
    -1.48
    ":[
    -1.46
    ĥ½
    -1.44
     compared
    -1.44
    ¦
    -1.43
    [{
    -1.42
    ¹
    -1.42
    ı
    -1.41
    POSITIVE LOGITS
    ptions
    2.24
     filings
    1.79
    lement
    1.77
    ctions
    1.74
    ious
    1.71
    sworth
    1.66
    ycin
    1.63
     protections
    1.62
    arial
    1.59
    ña
    1.59
    Activations Density 0.101%

    No Known Activations