© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. MLP Post
    5. 0-MLP-SM
    6. 28076
    Prev
    Next
    INDEX
    Explanations

    instances of the word "scintillating."

    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
    111
    +0.16
    0.9%
    148
    +0.11
    0.6%
    315
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    111
    +0.16
    0.01
    83
    +0.11
    0.01
    189
    +0.10
    0.01
    Negative Logits
    wards
    -1.71
    rapeut
    -1.60
    orems
    -1.55
     respondent
    -1.48
    shore
    -1.46
     Redist
    -1.44
    ments
    -1.43
    efore
    -1.43
    à³
    -1.38
     complaining
    -1.38
    POSITIVE LOGITS
    aho
    1.75
    illin
    1.68
    istor
    1.67
     embargo
    1.64
    igan
    1.63
    illac
    1.62
    illation
    1.61
    illa
    1.60
    oon
    1.60
    olta
    1.58
    Activations Density 0.703%

    No Known Activations