© 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. 24228
    Prev
    Next
    INDEX
    Explanations

    HTML tags or elements

    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
    156
    +0.14
    0.8%
    13
    +0.13
    0.7%
    47
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    13
    +0.14
    0.04
    332
    +0.13
    0.04
    109
    +0.12
    0.04
    Negative Logits
    reek
    -1.42
     resolution
    -1.42
    Permission
    -1.41
     minimization
    -1.38
     ?
    -1.35
     %.
    -1.34
    pace
    -1.34
    rvert
    -1.33
    ľĵ
    -1.32
     meeting
    -1.31
    POSITIVE LOGITS
    ^âĪĴ
    1.77
    )--
    1.53
    ducer
    1.51
    udson
    1.50
    brit
    1.47
     arriv
    1.41
    \]-
    1.39
    nar
    1.38
     than
    1.36
    nen
    1.36
    Activations Density 0.144%

    No Known Activations