© 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. 2-RES-SM
    11. 13025
    Prev
    Next
    INDEX
    Explanations

    references to scientific theories, particularly in the context of physics and its implications

    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/2-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.2.hook_resid_post
    Hook Layer
    2
    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
    189
    +0.16
    0.9%
    352
    +0.16
    0.9%
    356
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    189
    +0.16
    0.03
    8
    +0.16
    0.02
    352
    +0.14
    0.02
    Negative Logits
    µ
    -2.52
    ¢
    -2.42
    ¬
    -2.37
    ¿
    -2.32
    »
    -2.28
    §
    -2.20
    ª
    -2.19
    ¼
    -2.16
    ĥ
    -2.11
    ½
    -2.09
    POSITIVE LOGITS
    atics
    2.30
     cosmology
    1.86
     relativity
    1.77
    ical
    1.76
    lor
    1.73
     predicts
    1.70
    ically
    1.68
    atically
    1.65
    advanced
    1.60
    imetry
    1.59
    Activations Density 0.159%

    No Known Activations