© 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. 27810
    Prev
    Next
    INDEX
    Explanations

    technical medical terms and conditions related to congenital heart defects and procedures

    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
    18
    +0.12
    0.7%
    271
    +0.12
    0.7%
    47
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    18
    +0.12
    0.06
    164
    +0.12
    0.05
    134
    +0.11
    0.04
    Negative Logits
     purposes
    -1.73
     yours
    -1.69
     ingredient
    -1.68
     ingredients
    -1.66
     purpose
    -1.62
     behalf
    -1.61
    blogger
    -1.60
     sake
    -1.60
     theirs
    -1.49
     hiring
    -1.49
    POSITIVE LOGITS
    ³
    3.01
    ŀ
    2.75
    ģ
    2.72
    ¯
    2.68
    ħ
    2.67
    °
    2.64
    Ħ
    2.60
    ±
    2.58
    Ŀ
    2.55
    ²
    2.54
    Activations Density 0.343%

    No Known Activations