© 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. MLP Post
    9. 2-MLP-SM
    11. 25721
    Prev
    Next
    INDEX
    Explanations

    phrases indicating various forms of necessity or obligation

    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/2-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.2.hook_mlp_out
    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
    293
    +0.14
    0.8%
    99
    +0.14
    0.8%
    353
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    293
    +0.14
    0.06
    94
    +0.14
    0.06
    218
    +0.13
    0.05
    Negative Logits
    Ī
    -2.79
    -2.69
    <|outofrange|>
    -2.69
    -2.69
    -2.69
    ↵
    -2.69
    ↵
    -2.69
     ↵
    -2.69
    ↵
    -2.69
    <|outofrange|>
    -2.69
    POSITIVE LOGITS
    application
    1.64
     spot
    1.59
    icult
    1.47
     area
    1.42
    endas
    1.34
    akh
    1.33
    Ph
    1.33
     ISO
    1.32
     Torres
    1.31
    paces
    1.27
    Activations Density 1.644%

    No Known Activations