© 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. 32289
    Prev
    Next
    INDEX
    Explanations

    research studies and their associated findings

    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
    159
    +0.14
    0.8%
    235
    +0.14
    0.8%
    189
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    343
    +0.14
    0.03
    189
    +0.14
    0.03
    127
    +0.11
    0.03
    Negative Logits
    ĸ
    -3.38
    ¬
    -3.38
    §
    -3.34
    ĨĴ
    -3.32
    ¶
    -3.26
    Īĺ
    -3.21
    Ķ
    -3.17
    Ļª
    -3.16
    ¸
    -3.15
    ŀ
    -3.09
    POSITIVE LOGITS
     respects
    1.66
     aims
    1.56
     firstly
    1.52
    uminous
    1.50
     extends
    1.49
    autiful
    1.48
     ende
    1.41
    ariate
    1.38
     comprises
    1.38
     includes
    1.38
    Activations Density 0.106%

    No Known Activations