© 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. 14878
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "Fort."

    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
    445
    +0.13
    0.7%
    224
    +0.11
    0.6%
    369
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    467
    +0.13
    0.01
    373
    +0.11
    0.02
    410
    +0.11
    0.02
    Negative Logits
     lifetime
    -1.91
     subgroup
    -1.67
    ptions
    -1.62
    ergic
    -1.61
    ogene
    -1.60
    helf
    -1.57
     generation
    -1.50
    arner
    -1.48
     reticulum
    -1.43
     insertions
    -1.42
    POSITIVE LOGITS
    ress
    2.27
     Worth
    2.25
    night
    2.19
    itude
    2.08
    ifications
    2.06
    resses
    2.04
    ified
    1.90
    esc
    1.89
    ify
    1.85
     Knox
    1.84
    Activations Density 0.152%

    No Known Activations