© 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. 13335
    Prev
    Next
    INDEX
    Explanations

    references to JSON operations, particularly involving object manipulations and logging

    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
    21
    +0.33
    1.9%
    432
    +0.13
    0.7%
    263
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    432
    +0.33
    0.06
    469
    +0.13
    0.05
    110
    +0.10
    0.04
    Negative Logits
    »¿
    -2.49
    engers
    -2.00
    ¦
    -1.98
    iers
    -1.95
    uation
    -1.93
    uating
    -1.78
    ers
    -1.74
    edo
    -1.72
    Ŀ
    -1.69
    Ľ
    -1.67
    POSITIVE LOGITS
     undefined
    1.53
    printStackTrace
    1.48
     latitude
    1.38
     temp
    1.38
    stderr
    1.36
     guess
    1.35
     nowhere
    1.34
     elegans
    1.34
     starving
    1.33
     fatal
    1.32
    Activations Density 0.227%

    No Known Activations