© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 32476
    Prev
    Next
    INDEX
    Explanations

    phrases related to user interaction and customer engagement

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Neuron Alignment
    Index
    Value
    % of L₁
    419
    +0.22
    1.2%
    263
    +0.21
    1.2%
    186
    +0.18
    1.0%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    186
    +0.22
    0.06
    263
    +0.21
    0.09
    240
    +0.18
    -0.00
    Negative Logits
    Īĺ
    -2.96
    ¿½
    -2.71
    ±
    -2.42
    °
    -2.38
    ĥ½
    -2.35
    č↵
    -2.35
    č↵
    -2.35
    -2.35
    ↵  ³³³
    -2.35
    ↵
    -2.35
    POSITIVE LOGITS
    unov
    1.85
     detection
    1.66
     access
    1.63
     intellig
    1.61
     listener
    1.61
    win
    1.61
     assign
    1.59
    za
    1.52
    lez
    1.52
    /+
    1.49
    Activations Density 1.299%

    No Known Activations