© 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. 4-RES-SM
    11. 10482
    Prev
    Next
    INDEX
    Explanations

    technical terms and parameters related to coding and programming environments

    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/4-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.4.hook_resid_post
    Hook Layer
    4
    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
    +0.21
    1.2%
    95
    +0.12
    0.7%
    253
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    436
    +0.21
    0.04
    253
    +0.12
    0.07
    170
    +0.12
    0.04
    Negative Logits
     preponder
    -1.77
    acious
    -1.70
     economically
    -1.68
    ]{.
    -1.65
    aternal
    -1.60
    bags
    -1.58
    ALTH
    -1.56
    [**
    -1.54
    ^\[[@
    -1.51
     annually
    -1.50
    POSITIVE LOGITS
    ¥
    2.82
     code
    2.67
     version
    2.59
    ķ
    2.49
    Ł
    2.37
    ¿
    2.30
     ones
    2.28
     server
    2.27
     javascript
    2.27
     HTML
    2.22
    Activations Density 0.576%

    No Known Activations