© 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. 13220
    Prev
    Next
    INDEX
    Explanations

    legal terminology and references to court cases or opinions

    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
    12
    +0.13
    0.7%
    98
    +0.13
    0.7%
    329
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    329
    +0.13
    0.04
    253
    +0.13
    0.05
    391
    +0.11
    0.05
    Negative Logits
     etc
    -1.69
    .),
    -1.58
     +=
    -1.57
    qq
    -1.49
    typeof
    -1.47
    while
    -1.47
    conversion
    -1.42
    -)
    -1.41
    ...)
    -1.40
    print
    -1.38
    POSITIVE LOGITS
    ´
    1.86
    Ĥ¬
    1.82
    ¸
    1.80
    ł
    1.71
    ģ
    1.69
    ±
    1.68
    ¯
    1.64
    ¾
    1.63
    ŀ
    1.59
    °
    1.58
    Activations Density 0.314%

    No Known Activations