© 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. 27327
    Prev
    Next
    INDEX
    Explanations

    phrases related to job applications and qualifications

    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
    313
    +0.13
    0.8%
    450
    +0.13
    0.8%
    212
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    450
    +0.13
    0.10
    251
    +0.13
    0.07
    364
    +0.13
    0.10
    Negative Logits
    áĢ¬
    -1.74
    áĢº
    -1.62
     lover
    -1.52
    àµį
    -1.51
     lovers
    -1.47
    à«ĩ
    -1.46
     _(
    -1.43
    (_
    -1.41
     listener
    -1.40
    áĢ¯
    -1.40
    POSITIVE LOGITS
    gin
    1.55
    etin
    1.52
     Spring
    1.39
     Draft
    1.37
    stown
    1.34
    agin
    1.34
     tomorrow
    1.34
     lottery
    1.33
     next
    1.31
    cember
    1.30
    Activations Density 4.924%

    No Known Activations