© 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. 12770
    Prev
    Next
    INDEX
    Explanations

    programming-related keywords and syntax elements

    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
    90
    +0.13
    0.7%
    136
    +0.13
    0.7%
    198
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    53
    +0.13
    0.04
    209
    +0.13
    0.04
    90
    +0.12
    0.03
    Negative Logits
    essen
    -1.55
    .[]{
    -1.50
    imon
    -1.47
    ocy
    -1.46
    omatic
    -1.43
    hered
    -1.42
     gold
    -1.41
    ↵Č
    -1.41
     """
    -1.41
    ↵↵ ↵
    -1.40
    POSITIVE LOGITS
    ĭ
    3.64
    ·
    3.61
    ĻĤ
    3.46
    ¸
    3.42
    »
    3.42
    º
    3.39
    Ģ
    3.35
    į
    3.33
    ±
    3.31
    Ŀ
    3.31
    Activations Density 0.239%

    No Known Activations