© 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. 30792
    Prev
    Next
    INDEX
    Explanations

    elements related to programming or coding terminology

    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
    395
    +0.16
    0.9%
    510
    +0.13
    0.7%
    499
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    395
    +0.16
    0.13
    510
    +0.13
    0.14
    306
    +0.13
    0.09
    Negative Logits
    ¦
    -2.22
    ®
    -1.82
    «
    -1.72
    ¸
    -1.71
    Ń
    -1.66
    ĭ
    -1.64
    ¨
    -1.62
    ¼
    -1.59
    ª
    -1.58
    ħ
    -1.56
    POSITIVE LOGITS
     pulp
    1.60
     tongues
    1.42
    til
    1.31
     chances
    1.28
     depth
    1.24
     distance
    1.23
     advantage
    1.22
     interpolation
    1.22
    tered
    1.17
     successive
    1.16
    Activations Density 4.239%

    No Known Activations