© 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. 9522
    Prev
    Next
    INDEX
    Explanations

    numerical or statistical data within the text

    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
    +0.29
    1.8%
    89
    +0.16
    1.0%
    25
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    111
    +0.29
    0.41
    156
    +0.16
    0.38
    71
    +0.12
    0.28
    Negative Logits
    ĻĤ
    -5.75
    §
    -5.70
    -5.67
    -5.67
     ↵
    -5.67
    ↵
    -5.67
    -5.67
    ↵
    -5.67
    ↵
    -5.67
    ↵↵
    -5.67
    POSITIVE LOGITS
     even
    1.46
    plicity
    1.33
    gens
    1.31
    dling
    1.27
     respectively
    1.26
    cing
    1.25
    ches
    1.22
     countless
    1.22
    cheon
    1.22
    soever
    1.21
    Activations Density 4.869%

    No Known Activations