© 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. 25430
    Prev
    Next
    INDEX
    Explanations

    structural elements related to programming and object-oriented concepts

    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
    285
    +0.14
    0.8%
    369
    +0.12
    0.7%
    300
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    277
    +0.14
    0.04
    285
    +0.12
    0.04
    129
    +0.11
    0.01
    Negative Logits
    ģ
    -2.04
    atern
    -1.80
    ĻĤ
    -1.80
    «
    -1.68
    ²
    -1.67
    ĵ
    -1.67
    ĥ½
    -1.65
    ĩ
    -1.59
    ptin
    -1.57
    ontal
    -1.54
    POSITIVE LOGITS
    ed
    2.40
    ede
    2.23
    ing
    2.15
    edly
    1.69
    yard
    1.66
    edi
    1.66
    esan
    1.60
    s
    1.57
    merce
    1.51
    talk
    1.49
    Activations Density 0.257%

    No Known Activations