© 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. 14248
    Prev
    Next
    INDEX
    Explanations

    date and time 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
    78
    +0.17
    1.0%
    21
    +0.15
    0.9%
    158
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    158
    +0.17
    0.04
    162
    +0.15
    0.02
    172
    +0.14
    0.04
    Negative Logits
    ifice
    -1.77
    iciary
    -1.64
    soever
    -1.58
    ique
    -1.56
    gery
    -1.56
    iscus
    -1.53
    hers
    -1.51
    iques
    -1.47
     excel
    -1.47
    also
    -1.43
    POSITIVE LOGITS
    ©
    2.29
    Ĩ
    2.28
    ĥ
    2.27
     UTC
    2.27
    ¿
    2.16
    ļ
    2.15
    į
    2.12
    »
    2.07
    ĭ
    2.05
    ı
    2.05
    Activations Density 0.189%

    No Known Activations