© 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 EvalsExportsinterp-engineNEWAPI 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. 24621
    Prev
    Next
    INDEX
    Explanations

    phrases indicating the passage of time, particularly with the word "On" followed by a date

    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
    +0.16
    0.9%
    376
    +0.14
    0.8%
    243
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    243
    +0.16
    0.05
    496
    +0.14
    0.04
    57
    +0.13
    0.04
    Negative Logits
    illery
    -1.73
    icular
    -1.60
     variables
    -1.60
    schemas
    -1.49
     antigens
    -1.46
     %>
    -1.46
     tags
    -1.46
     biases
    -1.46
     Pradesh
    -1.42
     away
    -1.40
    POSITIVE LOGITS
    ¿½
    3.50
    «
    3.10
                ↵
    2.97
    ↵
    2.97
    Â
    2.97
    <|outofrange|>
    2.97
    2.97
    2.97
    2.97
    2.97
    Activations Density 0.266%

    No Known Activations