© 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. 23585
    Prev
    Next
    INDEX
    Explanations

    words related to undergarments

    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
    162
    +0.13
    0.7%
    459
    +0.11
    0.6%
    23
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    162
    +0.13
    0.08
    23
    +0.11
    0.03
    431
    +0.10
    0.03
    Negative Logits
    iates
    -1.67
    ification
    -1.64
    rais
    -1.57
    för
    -1.52
     ty
    -1.52
    craft
    -1.42
    ercise
    -1.40
    pile
    -1.38
    iation
    -1.37
    raiser
    -1.37
    POSITIVE LOGITS
    ĥ½
    1.70
    £
    1.57
    beam
    1.44
     Biosystems
    1.44
    aeda
    1.38
    '">
    1.37
    hin
    1.37
    inia
    1.36
    achi
    1.35
     0000000000000000000000000000000000
    1.32
    Activations Density 0.823%

    No Known Activations