© 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. 1-RES-SM
    11. 26459
    Prev
    Next
    INDEX
    Explanations

    generic phrases indicating existence or presence

    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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    23
    +0.15
    0.8%
    478
    +0.15
    0.8%
    491
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    68
    +0.15
    0.05
    491
    +0.15
    0.04
    9
    +0.12
    0.02
    Negative Logits
    ĭ
    -3.06
    Ī
    -2.96
    Ĭ
    -2.96
    Ļª
    -2.84
    »¿
    -2.75
    Ĵ
    -2.73
    ĩ
    -2.72
    ī
    -2.69
    İ
    -2.69
    ¶
    -2.66
    POSITIVE LOGITS
    spir
    1.61
    cludes
    1.60
     hope
    1.58
     Caption
    1.47
     sut
    1.46
     place
    1.44
    bbe
    1.44
    irections
    1.43
     brave
    1.43
     everything
    1.43
    Activations Density 0.323%

    No Known Activations