© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. MLP Post
    9. 0-MLP-SM
    11. 29092
    Prev
    Next
    INDEX
    Explanations

    instances of the word "correspond" and its variations, indicating connections or relationships between elements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__mlp-sm_processed/0-mlp-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.0.hook_mlp_out
    Hook Layer
    0
    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
    376
    +0.17
    1.0%
    507
    +0.15
    0.8%
    72
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    347
    +0.17
    0.01
    72
    +0.15
    0.01
    507
    +0.14
    0.01
    Negative Logits
     reputation
    -1.55
    ´
    -1.51
    Ħ
    -1.46
    ·¸
    -1.40
    ichi
    -1.36
    Appeal
    -1.36
    >();
    -1.34
    MO
    -1.33
     conscience
    -1.32
    umin
    -1.31
    POSITIVE LOGITS
    antes
    1.94
    ently
    1.92
    ents
    1.91
    imal
    1.77
    entes
    1.70
    itet
    1.69
    encia
    1.68
    iale
    1.67
    ences
    1.66
    enza
    1.64
    Activations Density 0.009%

    No Known Activations