© 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. 1-RES-SM
    11. 29825
    Prev
    Next
    INDEX
    Explanations

    references to electric-related concepts and technologies

    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
    156
    +0.18
    1.0%
    2
    +0.14
    0.8%
    225
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    331
    +0.18
    0.02
    225
    +0.14
    0.02
    2
    +0.14
    0.02
    Negative Logits
     hers
    -1.87
     apologize
    -1.75
     remorse
    -1.62
    ego
    -1.61
    atal
    -1.57
     reasoned
    -1.55
     condem
    -1.51
     retreat
    -1.51
     confess
    -1.50
    owing
    -1.50
    POSITIVE LOGITS
    Ł
    1.97
    ¬
    1.93
    ¦
    1.68
    Ĵ
    1.65
    volt
    1.63
    ķ
    1.58
    ŀ
    1.56
    ¥
    1.53
    ĸ
    1.51
    iens
    1.50
    Activations Density 0.046%

    No Known Activations