© 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. MLP Post
    9. 2-MLP-SM
    11. 22954
    Prev
    Next
    INDEX
    Explanations

    the word "that" in various contexts

    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
    455
    +0.11
    0.6%
    44
    +0.11
    0.6%
    93
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    502
    +0.11
    0.10
    44
    +0.11
    0.04
    403
    +0.11
    0.06
    Negative Logits
     progress
    -1.55
     sow
    -1.45
    yland
    -1.42
    orie
    -1.37
    ads
    -1.36
     cease
    -1.36
     retrospect
    -1.36
    imes
    -1.35
     older
    -1.34
     growth
    -1.34
    POSITIVE LOGITS
    Ļª
    2.63
    ello
    2.11
    ukin
    1.95
    ĩ
    1.76
    heet
    1.76
    llo
    1.72
    ĻĤ
    1.71
    ı
    1.71
    ŀ
    1.70
    ĨĴ
    1.64
    Activations Density 0.844%

    No Known Activations