© 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. 9341
    Prev
    Next
    INDEX
    Explanations

    slashes in 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
    347
    +0.14
    0.8%
    255
    +0.14
    0.8%
    21
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    347
    +0.14
    0.03
    452
    +0.14
    0.02
    436
    +0.13
    0.02
    Negative Logits
    ingly
    -1.72
    ically
    -1.67
     like
    -1.60
     enough
    -1.55
    lessly
    -1.51
     sense
    -1.48
    izing
    -1.46
     limits
    -1.45
    icular
    -1.41
     proportions
    -1.40
    POSITIVE LOGITS
    ı
    2.05
    Ŀ
    1.82
    ĭ
    1.81
    à°¿
    1.74
    à±
    1.70
    Ĩ
    1.70
    à±ģ
    1.70
    ĥ½
    1.63
    idegger
    1.59
    âķIJâķIJâķIJâķIJ
    1.59
    Activations Density 0.085%

    No Known Activations