© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 11116
    Prev
    Next
    INDEX
    Explanations

    instances of legal cases and citations

    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
    410
    +0.20
    1.1%
    490
    +0.14
    0.8%
    241
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    410
    +0.20
    0.03
    55
    +0.14
    0.02
    241
    +0.13
    0.02
    Negative Logits
    \[[@
    -1.55
     ?
    -1.53
    ()),
    -1.48
     knows
    -1.45
     <-
    -1.44
     converter
    -1.44
    )",
    -1.43
    frontal
    -1.42
    optera
    -1.39
     functioning
    -1.39
    POSITIVE LOGITS
    oted
    1.69
     Admiral
    1.61
    oting
    1.57
     aloud
    1.46
    ailed
    1.43
    ably
    1.43
     Colonel
    1.43
    Brad
    1.42
    Appeal
    1.40
    aternal
    1.37
    Activations Density 0.043%

    No Known Activations