© 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. 26083
    Prev
    Next
    INDEX
    Explanations

    words or phrases indicating significant performance or system errors

    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
    419
    +0.14
    0.8%
    302
    +0.13
    0.7%
    263
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    154
    +0.14
    0.03
    277
    +0.13
    0.02
    312
    +0.12
    0.02
    Negative Logits
    engers
    -1.93
    iast
    -1.74
    imon
    -1.71
    emen
    -1.57
    ener
    -1.55
    icks
    -1.54
    ees
    -1.53
    imes
    -1.52
    enders
    -1.47
    isters
    -1.45
    POSITIVE LOGITS
    ––––––––
    1.57
    dered
    1.53
    "...
    1.42
    §
    1.42
    etheless
    1.40
    -------------------------------------------------------
    1.28
    loat
    1.28
    »
    1.28
    ------------------------------------------------------
    1.27
    .....
    1.27
    Activations Density 0.129%

    No Known Activations