© 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. 0-MLP-SM
    11. 30147
    Prev
    Next
    INDEX
    Explanations

    instances of the word "saying"

    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
    +0.20
    1.1%
    32
    +0.12
    0.7%
    115
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    32
    +0.20
    0.03
    113
    +0.12
    0.02
    441
    +0.12
    0.02
    Negative Logits
    »¿
    -3.17
    »
    -3.05
    §
    -3.03
    ¡
    -3.00
    ·¸
    -2.94
    ŀ
    -2.90
     ↵
    -2.82
    -2.82
    -2.82
    ↵
    -2.82
    POSITIVE LOGITS
     thick
    1.53
     proper
    1.48
     order
    1.43
     groove
    1.43
     powerful
    1.40
     Confederate
    1.40
     superiority
    1.40
     thinner
    1.39
     decent
    1.39
     sickness
    1.36
    Activations Density 0.145%

    No Known Activations