© 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. 31296
    Prev
    Next
    INDEX
    Explanations

    occurrences of the verb "had" 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/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
    23
    +0.15
    0.9%
    410
    +0.14
    0.8%
    478
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    247
    +0.15
    0.06
    356
    +0.14
    0.06
    304
    +0.14
    0.05
    Negative Logits
    Ļª
    -3.20
    ł
    -2.59
    »¿
    -2.41
    ĭ
    -2.29
    Ģ
    -2.26
    ľ
    -2.22
    ģ
    -2.21
    Ń
    -2.20
    ŀ
    -2.19
    £
    -2.18
    POSITIVE LOGITS
    dock
    2.47
    oop
    2.38
    been
    1.95
     been
    1.89
    rien
    1.77
     fallen
    1.74
     recently
    1.73
     previously
    1.64
    ipes
    1.63
     unsuccess
    1.63
    Activations Density 0.158%

    No Known Activations