© 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. 9289
    Prev
    Next
    INDEX
    Explanations

    numerical values and mathematical terms

    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
    214
    +0.12
    0.7%
    267
    +0.11
    0.7%
    78
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    267
    +0.12
    0.02
    214
    +0.11
    0.01
    436
    +0.11
    0.01
    Negative Logits
    ly
    -1.68
    tti
    -1.49
    ::
    -1.49
    ships
    -1.44
    ken
    -1.41
    past
    -1.41
    ing
    -1.38
    idge
    -1.37
    leneck
    -1.35
    bing
    -1.34
    POSITIVE LOGITS
    Ļ
    1.79
    ·
    1.78
     prob
    1.69
    Īĺ
    1.59
     thou
    1.55
    ħ
    1.38
    οÏħμε
    1.34
    ĺ
    1.33
    £
    1.31
    ĵ
    1.27
    Activations Density 0.026%

    No Known Activations