© 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. 27605
    Prev
    Next
    INDEX
    Explanations

    references to specific years, particularly 1991 and surrounding years

    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
    308
    +0.15
    0.8%
    317
    +0.13
    0.7%
    333
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    308
    +0.15
    0.04
    475
    +0.13
    0.03
    51
    +0.12
    0.03
    Negative Logits
    ĥ
    -3.30
    ¿
    -3.29
    ¼
    -3.25
    Ħ
    -3.09
    º
    -2.96
    Ļ
    -2.94
    ¸
    -2.93
    Ĥ¬
    -2.88
    ī
    -2.85
    ·
    -2.84
    POSITIVE LOGITS
     anniversary
    1.86
    icolor
    1.76
    nai
    1.68
    áĢ¸
    1.62
    áĢ¬
    1.60
    ITED
    1.60
    odia
    1.57
    à¯į
    1.54
    à±į
    1.53
    olecular
    1.51
    Activations Density 0.045%

    No Known Activations