© 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. 28834
    Prev
    Next
    INDEX
    Explanations

    multiple mentions of the word "and."

    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
    272
    +0.17
    1.0%
    39
    +0.12
    0.7%
    341
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    494
    +0.17
    0.45
    209
    +0.12
    0.29
    402
    +0.10
    0.26
    Negative Logits
     sickness
    -1.54
    etime
    -1.46
    shot
    -1.44
    nier
    -1.40
    ate
    -1.39
    ked
    -1.36
    sects
    -1.35
    .");
    -1.32
    hell
    -1.31
    !");
    -1.31
    POSITIVE LOGITS
    °
    1.77
    ¿
    1.64
    encial
    1.49
    ĻĤ
    1.47
    İ
    1.45
    crets
    1.41
     meant
    1.38
    ISPR
    1.36
     lig
    1.36
    ération
    1.36
    Activations Density 1.562%

    No Known Activations