© 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. 15286
    Prev
    Next
    INDEX
    Explanations

    references to software licenses

    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
    78
    +0.22
    1.3%
    391
    +0.11
    0.6%
    271
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    241
    +0.22
    0.01
    157
    +0.11
    0.01
    368
    +0.11
    0.01
    Negative Logits
    ļ
    -3.56
    ĥ½
    -3.43
    ľ
    -3.17
    Ģ
    -3.17
    Ń
    -3.12
    ĸ´
    -3.09
    ľĵ
    -3.05
    ķ
    -3.05
    ¤
    -3.02
    ı
    -2.99
    POSITIVE LOGITS
    uego
    1.40
     validity
    1.33
    Ïģε
    1.29
     receipt
    1.28
    unto
    1.27
    μα
    1.25
    raged
    1.25
    yer
    1.25
    лÑı
    1.24
     agreement
    1.24
    Activations Density 0.041%

    No Known Activations