© 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. 4-RES-SM
    11. 13802
    Prev
    Next
    INDEX
    Explanations

    technical or programming-related syntax or structures

    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
    78
    +0.14
    0.8%
    75
    +0.11
    0.6%
    432
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    78
    +0.14
    -0.00
    397
    +0.11
    0.04
    506
    +0.11
    0.01
    Negative Logits
     Amy
    -1.94
    idis
    -1.70
     Mead
    -1.58
     NJ
    -1.53
     Ford
    -1.52
    enu
    -1.48
    wagen
    -1.45
    termination
    -1.43
    itter
    -1.41
    isy
    -1.38
    POSITIVE LOGITS
    8
    2.27
     8
    2.17
     888
    1.89
    832
    1.79
    812
    1.78
    833
    1.77
    896
    1.77
    816
    1.72
     820
    1.70
    818
    1.67
    Activations Density 0.718%

    No Known Activations