© 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. 14037
    Prev
    Next
    INDEX
    Explanations

    references to the letter "J" or related character sequences

    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
    369
    +0.15
    0.8%
    172
    +0.13
    0.7%
    204
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    369
    +0.15
    0.04
    172
    +0.13
    0.04
    204
    +0.13
    0.04
    Negative Logits
    pective
    -1.82
    ting
    -1.67
    rers
    -1.63
    ging
    -1.62
    cted
    -1.56
    king
    -1.56
    acles
    -1.56
    aneously
    -1.54
    cher
    -1.52
     otherwise
    -1.50
    POSITIVE LOGITS
    unction
    2.67
    Ĥ¬
    2.55
    ¶
    2.38
    Ģ
    2.36
    ĨĴ
    2.33
    IJ
    2.32
    Ĩ
    2.23
    ¬
    2.21
    £
    2.16
    ĺ
    2.15
    Activations Density 0.055%

    No Known Activations