© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 22559
    Prev
    Next
    INDEX
    Explanations

    instances of the word "However."

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Neuron Alignment
    Index
    Value
    % of L₁
    475
    +0.15
    0.8%
    332
    +0.12
    0.6%
    98
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    109
    +0.15
    0.06
    479
    +0.12
    0.04
    475
    +0.12
    0.05
    Negative Logits
    iNdEx
    -1.75
    autre
    -1.46
     invent
    -1.44
    liv
    -1.39
    à±į
    -1.38
    quad
    -1.37
     okay
    -1.37
    ophyll
    -1.35
    fixed
    -1.34
     satisf
    -1.34
    POSITIVE LOGITS
    ³
    2.49
    ·¸
    2.45
    ĨĴ
    2.28
    »
    2.17
    ŀ
    2.17
    ĸ´
    2.14
    ĥ½
    2.10
    ¶
    2.08
    Ĭ
    2.05
    Ĩ
    2.01
    Activations Density 0.087%

    No Known Activations