© 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. 29383
    Prev
    Next
    INDEX
    Explanations

    expressions of positive emotions and welcoming sentiments

    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
    346
    +0.12
    0.7%
    65
    +0.12
    0.6%
    402
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    346
    +0.12
    0.03
    402
    +0.12
    0.04
    505
    +0.12
    0.04
    Negative Logits
    ¯
    -2.06
    º
    -2.01
    Ģ
    -1.87
    ¬
    -1.75
    ozo
    -1.65
    )\].
    -1.65
    illin
    -1.62
    Īĺ
    -1.62
    ¹
    -1.61
    į
    -1.55
    POSITIVE LOGITS
     shiny
    1.50
     loyal
    1.48
     welcoming
    1.45
     '
    1.45
     welcome
    1.44
     bright
    1.42
    standing
    1.41
    kered
    1.40
     experience
    1.40
    eting
    1.38
    Activations Density 0.827%

    No Known Activations