© 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. 27488
    Prev
    Next
    INDEX
    Explanations

    references to the JavaScript programming 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/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
    +0.28
    1.6%
    365
    +0.14
    0.8%
    82
    +0.11
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    111
    +0.28
    0.05
    271
    +0.14
    0.02
    156
    +0.11
    0.03
    Negative Logits
    ·
    -3.07
    ½
    -2.90
    »
    -2.82
    ¯
    -2.79
    Ĭ
    -2.77
    ł
    -2.75
    ķ
    -2.73
    ĺ
    -2.71
    ¿½
    -2.69
    ĥ½
    -2.69
    POSITIVE LOGITS
    blower
    1.70
    fonts
    1.64
    ulator
    1.49
    ulators
    1.49
    heses
    1.48
    scripts
    1.47
    cript
    1.47
     engine
    1.47
    Selector
    1.46
    heets
    1.46
    Activations Density 0.184%

    No Known Activations