© 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. 23801
    Prev
    Next
    INDEX
    Explanations

    references to the concept of "more" or increases in quantity or quality

    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
    241
    +0.13
    0.7%
    231
    +0.13
    0.7%
    125
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    125
    +0.13
    0.04
    241
    +0.13
    0.05
    18
    +0.12
    0.05
    Negative Logits
    ½
    -2.71
    ¼
    -2.62
    ¾
    -2.57
    ¹
    -2.55
    ·¸
    -2.36
    ľĵ
    -2.31
    ´
    -2.30
    ij
    -2.30
    ¶
    -2.25
    ĩ
    -2.20
    POSITIVE LOGITS
     than
    2.54
     forthcoming
    1.97
    Than
    1.95
     importantly
    1.79
     stringent
    1.71
    than
    1.71
    theless
    1.60
     sophisticated
    1.52
     Than
    1.51
     accurate
    1.48
    Activations Density 0.107%

    No Known Activations