© 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. 27793
    Prev
    Next
    INDEX
    Explanations

    repetitive use of the phrase "up"

    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
    363
    +0.16
    0.9%
    176
    +0.14
    0.8%
    319
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    12
    +0.16
    0.01
    363
    +0.14
    0.02
    273
    +0.13
    0.01
    Negative Logits
    ³
    -1.59
     mound
    -1.57
     vessel
    -1.54
    matically
    -1.49
     milestone
    -1.47
     Photograph
    -1.45
    Īĺ
    -1.44
    [_
    -1.44
    arer
    -1.40
    isor
    -1.40
    POSITIVE LOGITS
     juices
    1.52
    ashes
    1.50
     sentences
    1.49
    nis
    1.44
    nuts
    1.41
     lately
    1.41
     since
    1.40
     pitches
    1.40
     flavors
    1.36
     bugs
    1.36
    Activations Density 0.179%

    No Known Activations