© 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. 9666
    Prev
    Next
    INDEX
    Explanations

    text related to legal or formal notices, particularly copyright and licensing information

    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
    111
    +0.77
    14.9%
    156
    +0.58
    11.1%
    25
    +0.09
    1.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    111
    +0.77
    0.96
    156
    +0.58
    0.68
    71
    +0.09
    0.69
    Negative Logits
    ¿½
    -3.67
    -3.62
    ↵
    -3.62
    <|outofrange|>
    -3.62
    ↵
    -3.62
    ↵↵
    -3.62
    ↵↵
    -3.62
                ↵
    -3.62
     ↵
    -3.62
    ↵↵
    -3.62
    POSITIVE LOGITS
    rosis
    1.00
    rox
    0.95
    orph
    0.95
    olia
    0.93
     concentration
    0.92
     fundament
    0.91
     polÃŃ
    0.87
    roscopic
    0.86
     alone
    0.85
     um
    0.85
    Activations Density 1.841%

    No Known Activations