© 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. 13756
    Prev
    Next
    INDEX
    Explanations

    phrases related to data structure and processing in programming

    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
    436
    +0.14
    0.8%
    271
    +0.13
    0.8%
    6
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    43
    +0.14
    0.05
    351
    +0.13
    0.09
    122
    +0.13
    0.07
    Negative Logits
    ulent
    -1.50
     deserve
    -1.48
     designs
    -1.47
    ocur
    -1.39
    works
    -1.38
     influenced
    -1.34
     besides
    -1.33
    alore
    -1.33
     Architect
    -1.33
     owes
    -1.31
    POSITIVE LOGITS
    ¦
    1.71
    cribing
    1.68
     rhet
    1.61
    iginally
    1.60
    handed
    1.51
     "âĻª
    1.51
    heading
    1.51
     pause
    1.51
     disbelief
    1.50
     silence
    1.50
    Activations Density 4.234%

    No Known Activations