© 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 EvalsExportsAPI 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. 11540
    Prev
    Next
    INDEX
    Explanations

    the verb "be" and its variations in different 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/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
    218
    +0.16
    0.9%
    186
    +0.13
    0.7%
    307
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    218
    +0.16
    0.12
    144
    +0.13
    0.09
    272
    +0.13
    0.09
    Negative Logits
    Ħ
    -3.05
    Ī
    -3.02
    Ŀ
    -2.93
    ¥
    -2.89
    ī
    -2.88
    ª
    -2.86
    ľ
    -2.78
    ģ
    -2.69
    ķ
    -2.68
    Ĭ
    -2.67
    POSITIVE LOGITS
     able
    2.55
    friend
    2.34
     acquainted
    2.29
     aware
    2.22
     afraid
    2.18
     engaged
    2.18
     happier
    2.16
     married
    2.12
     invested
    2.11
     immersed
    2.10
    Activations Density 0.929%

    No Known Activations