© 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. 1-RES-SM
    11. 25706
    Prev
    Next
    INDEX
    Explanations

    occurrences of the term "us" in various contexts

    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/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
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    445
    +0.12
    0.7%
    343
    +0.10
    0.6%
    338
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    445
    +0.12
    0.03
    352
    +0.10
    0.03
    343
    +0.10
    0.03
    Negative Logits
    your
    -1.80
     your
    -1.65
    FROM
    -1.61
      ↵
    -1.58
     puff
    -1.52
     heartbeat
    -1.52
     ");
    -1.47
    FORE
    -1.47
    different
    -1.45
     sooner
    -1.45
    POSITIVE LOGITS
    cule
    2.20
    cript
    1.96
    heit
    1.90
    creen
    1.87
    olve
    1.85
    heets
    1.85
    burgh
    1.84
    pect
    1.81
    stadt
    1.80
    cules
    1.78
    Activations Density 0.100%

    No Known Activations