© 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. 1-RES-SM
    11. 29954
    Prev
    Next
    INDEX
    Explanations

    punctuation marks, particularly periods, in the text

    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/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
    Neuron Alignment
    Index
    Value
    % of L₁
    265
    +0.15
    0.8%
    56
    +0.12
    0.7%
    317
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    258
    +0.15
    0.00
    23
    +0.12
    0.00
    129
    +0.12
    0.00
    Negative Logits
    ĭ
    -2.82
    ¶
    -2.79
    µ
    -2.69
    »
    -2.67
    ¦
    -2.62
    ł
    -2.57
    ĵ
    -2.56
    ¾
    -2.53
    °
    -2.47
    Ģ
    -2.44
    POSITIVE LOGITS
    alus
    1.86
     jurisdiction
    1.84
    ruptcy
    1.71
    atories
    1.64
     me
    1.59
    etus
    1.58
    hei
    1.57
    als
    1.57
     Animals
    1.55
    achus
    1.53
    Activations Density 0.001%

    No Known Activations