© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 14222
    Prev
    Next
    INDEX
    Explanations

    technical terms and implementations related to programming and software development

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Neuron Alignment
    Index
    Value
    % of L₁
    287
    +0.15
    0.8%
    416
    +0.12
    0.7%
    17
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    17
    +0.15
    0.01
    416
    +0.12
    0.08
    157
    +0.12
    0.11
    Negative Logits
    nai
    -1.84
    ajo
    -1.60
    à«ĩ
    -1.50
    icken
    -1.49
     [âĢ¦]
    -1.48
    áĥĺ
    -1.43
    seys
    -1.42
    IGHT
    -1.40
    âĻª
    -1.36
    ีà¹Ī
    -1.36
    POSITIVE LOGITS
    ¿½
    1.70
     representation
    1.63
     Definition
    1.60
     initialization
    1.60
     Configuration
    1.59
    ģ
    1.58
     Structure
    1.54
     configuration
    1.51
     Character
    1.49
     Assert
    1.49
    Activations Density 1.857%

    No Known Activations