© 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. 14220
    Prev
    Next
    INDEX
    Explanations

    alternatives and options

    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
    316
    +0.12
    0.6%
    476
    +0.11
    0.6%
    472
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    316
    +0.12
    0.03
    503
    +0.11
    0.03
    200
    +0.11
    0.03
    Negative Logits
    urious
    -1.99
    urus
    -1.69
    ľĵ
    -1.60
    arser
    -1.58
    urized
    -1.56
    evol
    -1.55
    inence
    -1.54
    uries
    -1.53
    ured
    -1.53
    ês
    -1.49
    POSITIVE LOGITS
    bital
    2.58
    naments
    2.46
    chid
    2.37
    nament
    2.07
     equivalently
    2.07
    derer
    2.04
     something
    2.03
    ders
    2.00
     alternatively
    1.98
    phan
    1.96
    Activations Density 0.192%

    No Known Activations