© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 1-RES-SM
    11. 26244
    Prev
    Next
    INDEX
    Explanations

    the less than symbol "<"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__res-sm_processed/1-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.1.hook_resid_post
    Hook Layer
    1
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    478
    +0.23
    1.3%
    1
    +0.11
    0.6%
    397
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    124
    +0.23
    0.03
    103
    +0.11
    0.03
    444
    +0.10
    0.02
    Negative Logits
     ORDER
    -1.66
    aid
    -1.49
    tooltip
    -1.46
    isman
    -1.43
    aused
    -1.40
    illard
    -1.40
    otic
    -1.40
    aza
    -1.38
    izable
    -1.37
    ![](
    -1.37
    POSITIVE LOGITS
     another
    1.54
     ich
    1.52
     distant
    1.45
     miles
    1.45
     hypox
    1.37
     bathrooms
    1.35
     honest
    1.35
     hesitation
    1.34
     serm
    1.32
     dich
    1.32
    Activations Density 0.050%

    No Known Activations