© 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. 2-RES-SM
    11. 28627
    Prev
    Next
    INDEX
    Explanations

    references to the human body and its functions

    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/2-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.2.hook_resid_post
    Hook Layer
    2
    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
    356
    +0.16
    0.9%
    476
    +0.11
    0.6%
    407
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    356
    +0.16
    0.03
    93
    +0.11
    0.03
    128
    +0.11
    0.01
    Negative Logits
    ª
    -2.38
    ĩ
    -2.09
    Ī
    -1.91
    ĨĴ
    -1.91
    ľĵ
    -1.79
    §
    -1.78
    ĵ
    -1.77
     behalf
    -1.75
    Ĭ
    -1.72
    Ĵ
    -1.68
    POSITIVE LOGITS
    guard
    2.08
    guards
    2.03
    builder
    1.97
    weight
    1.88
    work
    1.83
    piece
    1.81
    bone
    1.75
    pieces
    1.70
     cavity
    1.69
    bones
    1.67
    Activations Density 0.101%

    No Known Activations