© 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. 4-RES-SM
    11. 15175
    Prev
    Next
    INDEX
    Explanations

    commands and actions related to requests and interactions

    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/4-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.4.hook_resid_post
    Hook Layer
    4
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Neuron Alignment
    Index
    Value
    % of L₁
    307
    +0.20
    1.1%
    303
    +0.12
    0.7%
    263
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    263
    +0.20
    0.15
    56
    +0.12
    0.06
    286
    +0.11
    0.13
    Negative Logits
    ·
    -3.56
    ļ
    -3.52
    ĩ
    -3.48
    ĵ
    -3.48
    ij
    -3.46
    ľĵ
    -3.46
    -3.36
         ↵
    -3.36
    ↵
    -3.36
    ↵
    -3.36
    POSITIVE LOGITS
     tomorrow
    2.63
     myself
    2.35
     someday
    2.28
     another
    2.11
     something
    2.05
     sooner
    2.00
     my
    2.00
     sometime
    1.93
     again
    1.72
     maybe
    1.72
    Activations Density 2.279%

    No Known Activations