© 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. 14164
    Prev
    Next
    INDEX
    Explanations

    phrases indicating direction or intention

    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
    69
    +0.13
    0.7%
    235
    +0.10
    0.6%
    439
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    69
    +0.13
    0.06
    500
    +0.10
    0.04
    190
    +0.10
    0.04
    Negative Logits
    ĻĤ
    -1.90
     sake
    -1.85
     directions
    -1.58
    ĸ
    -1.49
    cup
    -1.49
    ĩ
    -1.49
     perspective
    -1.44
    ificial
    -1.43
    ¦
    -1.42
    ª
    -1.41
    POSITIVE LOGITS
     endorse
    1.60
    CLAIM
    1.57
     expand
    1.57
     deprec
    1.56
     accumulate
    1.55
     penal
    1.54
     compensate
    1.54
     curb
    1.54
     prosecute
    1.54
     incarcer
    1.52
    Activations Density 0.286%

    No Known Activations