© 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. 1-RES-SM
    11. 24126
    Prev
    Next
    INDEX
    Explanations

    references to streets and urban environments

    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
    115
    +0.16
    0.9%
    156
    +0.14
    0.8%
    309
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    115
    +0.16
    0.02
    286
    +0.14
    0.02
    507
    +0.10
    0.01
    Negative Logits
    «
    -2.18
    ĥ
    -1.89
    Ī
    -1.88
    Ĩ
    -1.87
    ´
    -1.86
    ¥
    -1.86
    TRODUCTION
    -1.71
    º
    -1.71
    ±
    -1.67
     promise
    -1.65
    POSITIVE LOGITS
    cars
    2.10
    ways
    2.03
    wise
    1.87
    corner
    1.84
    works
    1.84
    holder
    1.79
     corners
    1.79
     scene
    1.68
    lights
    1.67
    floor
    1.62
    Activations Density 0.092%

    No Known Activations