© 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. 26712
    Prev
    Next
    INDEX
    Explanations

    phrases indicating comparative measurements or trends related to health, illness, or social statistics

    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
    451
    +0.13
    0.7%
    266
    +0.13
    0.7%
    457
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    457
    +0.13
    0.03
    266
    +0.13
    0.03
    451
    +0.12
    0.02
    Negative Logits
    ¼
    -2.56
    İ
    -2.50
    ¹
    -2.44
    ³
    -2.44
    ½
    -2.44
    ´
    -2.43
    ·¸
    -2.42
    º
    -2.36
    Į
    -2.32
    ¬
    -2.31
    POSITIVE LOGITS
     than
    3.59
    than
    2.95
    Than
    2.83
     Than
    2.49
     degrees
    2.27
     compared
    1.93
     magnitude
    1.88
     thresholds
    1.84
     levels
    1.84
     (>
    1.77
    Activations Density 0.066%

    No Known Activations