© 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. 13147
    Prev
    Next
    INDEX
    Explanations

    patterns of nested structures and related symbols in programming or data organization 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
    303
    +0.11
    0.6%
    44
    +0.11
    0.6%
    57
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    397
    +0.11
    0.05
    57
    +0.11
    0.05
    47
    +0.11
    0.04
    Negative Logits
    heed
    -1.76
    chini
    -1.65
    phia
    -1.57
     breakdown
    -1.51
    ruptcy
    -1.49
     Kelly
    -1.46
    rina
    -1.46
    Kelly
    -1.45
    iners
    -1.45
    apore
    -1.44
    POSITIVE LOGITS
     object
    3.60
     objects
    3.16
    objects
    3.11
    object
    3.00
     Object
    2.97
     Objects
    2.94
    Object
    2.86
    Objects
    2.80
    obj
    2.72
     obj
    2.69
    Activations Density 0.543%

    No Known Activations