© 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. 26157
    Prev
    Next
    INDEX
    Explanations

    names and references to individuals

    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
    369
    +0.20
    1.1%
    156
    +0.14
    0.8%
    484
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    369
    +0.20
    0.10
    410
    +0.14
    0.17
    67
    +0.13
    0.07
    Negative Logits
     accelerator
    -1.39
    ipes
    -1.34
     supported
    -1.33
    iano
    -1.29
    orption
    -1.29
    bras
    -1.29
    ipel
    -1.26
    bital
    -1.26
    /,
    -1.24
     [$
    -1.22
    POSITIVE LOGITS
     notice
    1.45
    ·¸
    1.42
     guess
    1.37
    arman
    1.35
    'm
    1.32
     glance
    1.30
     advise
    1.28
     discover
    1.28
     lies
    1.26
     glimpse
    1.21
    Activations Density 2.906%

    No Known Activations