© 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. 1-RES-SM
    11. 25736
    Prev
    Next
    INDEX
    Explanations

    mathematical questions and operations involving roots and powers of numbers

    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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    268
    +0.14
    0.8%
    329
    +0.13
    0.8%
    478
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    84
    +0.14
    0.01
    157
    +0.13
    0.03
    329
    +0.13
    0.02
    Negative Logits
    ®
    -1.92
     vs
    -1.81
    ĸ
    -1.72
    ļ
    -1.69
    ĭ
    -1.67
    works
    -1.63
     versus
    -1.62
    Ł
    -1.58
    §
    -1.57
    ģ
    -1.54
    POSITIVE LOGITS
    ureus
    2.11
     prob
    1.65
    century
    1.54
     hereby
    1.53
     WARRANTY
    1.53
    ircon
    1.48
    bbe
    1.48
    ÏģÏī
    1.47
    javase
    1.47
     periodontal
    1.39
    Activations Density 0.253%

    No Known Activations