© 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 EvalsExportsAPI 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. 28820
    Prev
    Next
    INDEX
    Explanations

    coding patterns related to ASCII encoding

    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
    159
    +0.14
    0.8%
    300
    +0.13
    0.7%
    380
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    53
    +0.14
    0.06
    438
    +0.13
    0.05
    56
    +0.13
    0.01
    Negative Logits
    ĥ½
    -1.66
     whole
    -1.66
     late
    -1.66
     him
    -1.66
    aning
    -1.64
     entire
    -1.63
     only
    -1.61
     missed
    -1.60
     distant
    -1.56
    eing
    -1.56
    POSITIVE LOGITS
    ULAR
    1.89
    ium
    1.81
    iere
    1.80
    II
    1.77
    mere
    1.70
    ADE
    1.67
    esan
    1.65
    hel
    1.65
    notes
    1.64
    etto
    1.64
    Activations Density 4.706%

    No Known Activations