© 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. 27346
    Prev
    Next
    INDEX
    Explanations

    titles and proper titles of individuals, emphasizing the name "Mr."

    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
    410
    +0.20
    1.1%
    156
    +0.18
    1.0%
    369
    +0.18
    1.0%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    410
    +0.20
    0.05
    177
    +0.18
    0.04
    369
    +0.18
    0.02
    Negative Logits
    Docket
    -1.91
    availability
    -1.62
    NOW
    -1.59
    oside
    -1.58
    break
    -1.51
    osition
    -1.47
    iencies
    -1.46
    ENTIAL
    -1.44
    OOGLE
    -1.40
    ){#
    -1.40
    POSITIVE LOGITS
     himself
    1.88
    ł
    1.77
    Į
    1.63
     forgiven
    1.55
     Stevens
    1.49
    ¿½
    1.45
    ¦
    1.45
     Biden
    1.44
     Morrison
    1.42
    Ĩ
    1.42
    Activations Density 0.143%

    No Known Activations