© 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. 1-RES-SM
    11. 24757
    Prev
    Next
    INDEX
    Explanations

    clusters of numerical calculations or operations

    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
    414
    +0.11
    0.6%
    102
    +0.11
    0.6%
    386
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    102
    +0.11
    0.02
    459
    +0.11
    0.02
    451
    +0.10
    0.01
    Negative Logits
    iginally
    -1.61
    cules
    -1.49
    uth
    -1.46
     district
    -1.45
    ubot
    -1.42
     supra
    -1.41
     abnormalities
    -1.40
    uits
    -1.40
    xture
    -1.39
     defects
    -1.38
    POSITIVE LOGITS
    ĥ
    3.60
    §
    3.33
    IJ
    3.22
    ¡
    3.19
    ¢
    3.16
    Ī
    3.13
    Ĩ
    3.01
    Ĵ
    2.94
    Ļ
    2.92
    2.85
    Activations Density 0.028%

    No Known Activations