© 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. 12716
    Prev
    Next
    INDEX
    Explanations

    dates and statistics related to events or actions

    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
    263
    +0.15
    0.9%
    98
    +0.13
    0.8%
    369
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    33
    +0.15
    0.04
    46
    +0.13
    0.02
    263
    +0.13
    0.04
    Negative Logits
    Īĺ
    -3.06
    ³
    -2.86
    ·¸
    -2.73
    ¹
    -2.49
    ¢
    -2.46
    ĻĤ
    -2.45
    Ļ
    -2.39
    ¿
    -2.38
    º
    -2.36
    £
    -2.35
    POSITIVE LOGITS
    arn
    1.50
    inai
    1.50
    fal
    1.43
    plastic
    1.42
    gon
    1.38
    wan
    1.37
    pected
    1.36
    80211
    1.35
    anus
    1.33
    itat
    1.32
    Activations Density 0.493%

    No Known Activations