© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 8664
    Prev
    Next
    INDEX
    Explanations

    programming or database commands and syntax

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Neuron Alignment
    Index
    Value
    % of L₁
    506
    +0.15
    0.9%
    46
    +0.13
    0.8%
    169
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    506
    +0.15
    0.02
    136
    +0.13
    -0.01
    30
    +0.12
    0.02
    Negative Logits
     IE
    -1.73
     Cave
    -1.44
     Iraqi
    -1.38
     GE
    -1.31
    boxes
    -1.31
    ż
    -1.31
    chle
    -1.30
     Biden
    -1.29
     WB
    -1.29
     Ba
    -1.28
    POSITIVE LOGITS
     16
    2.52
     15
    2.41
    16
    2.38
    15
    2.35
     14
    2.30
     17
    2.27
    17
    2.19
    1600
    2.09
    14
    2.06
     1600
    2.00
    Activations Density 0.503%

    No Known Activations