© 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. 25720
    Prev
    Next
    INDEX
    Explanations

    titles or references to individuals with the honorific "Sir."

    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
    410
    +0.18
    1.1%
    156
    +0.16
    0.9%
    226
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    226
    +0.18
    0.02
    410
    +0.16
    0.02
    270
    +0.13
    0.02
    Negative Logits
    ļ
    -1.62
    ĥ½
    -1.59
    ){#
    -1.56
    .""
    -1.53
    sterdam
    -1.40
    )."
    -1.38
    "?"
    -1.36
     corresponding
    -1.35
     separation
    -1.35
    "."
    -1.32
    POSITIVE LOGITS
    ius
    1.74
    pix
    1.68
    leaf
    1.53
    imates
    1.49
    croft
    1.47
    quito
    1.46
    kon
    1.46
    gments
    1.40
    culo
    1.39
    fileID
    1.39
    Activations Density 0.040%

    No Known Activations