© 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 EvalsExportsAPI 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. 11383
    Prev
    Next
    INDEX
    Explanations

    the word "it" and its occurrences in various contexts

    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
    21
    +0.15
    0.9%
    427
    +0.12
    0.7%
    441
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    427
    +0.15
    0.08
    441
    +0.12
    0.03
    26
    +0.12
    0.08
    Negative Logits
    Ļª
    -4.84
    ĻĤ
    -4.72
    Īĺ
    -3.96
    Ļ
    -3.96
    Ī
    -3.85
    Ģ
    -3.79
    Ŀ
    -3.77
    ľĵ
    -3.70
    ĥ
    -3.69
    Ĩ
    -3.68
    POSITIVE LOGITS
     happens
    2.51
     exists
    2.51
     pertains
    2.46
     appears
    2.39
     becomes
    2.32
    alian
    2.27
     comes
    2.26
     expires
    2.23
     occurs
    2.23
    anium
    2.18
    Activations Density 0.468%

    No Known Activations