© 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. 2-RES-SM
    11. 31098
    Prev
    Next
    INDEX
    Explanations

    references to the nursing profession and its associated concepts

    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/2-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.2.hook_resid_post
    Hook Layer
    2
    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
    156
    +0.21
    1.2%
    492
    +0.13
    0.7%
    230
    +0.11
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    230
    +0.21
    0.01
    418
    +0.13
    -0.00
    494
    +0.11
    0.01
    Negative Logits
     strangers
    -1.57
     narrowing
    -1.42
     stranger
    -1.41
     BGCOLOR
    -1.36
     recombination
    -1.35
     scrolling
    -1.32
     nesting
    -1.29
     saturated
    -1.26
    unden
    -1.26
    packed
    -1.25
    POSITIVE LOGITS
    maid
    2.14
    hurst
    1.88
    nai
    1.72
    ĩ
    1.71
    mable
    1.69
    imental
    1.68
    man
    1.61
    hall
    1.56
    mare
    1.54
    leneck
    1.50
    Activations Density 0.157%

    No Known Activations