Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. MLP Post
    5. 0-MLP-SM
    6. 27592
    Prev
    Next
    INDEX
    Explanations

    references to the actor Tom Cruise

    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
    276
    +0.12
    0.7%
    416
    +0.12
    0.6%
    503
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    276
    +0.12
    0.01
    416
    +0.12
    0.01
    383
    +0.11
    0.01
    Negative Logits
    ĥ½
    -1.94
    rapeut
    -1.69
    ¼
    -1.60
     death
    -1.58
    ulls
    -1.54
    ĻĤ
    -1.53
    plasia
    -1.51
     male
    -1.50
    mes
    -1.46
    ij
    -1.45
    POSITIVE LOGITS
    ulence
    1.50
    DOM
    1.45
    antry
    1.45
    dale
    1.43
    esp
    1.43
     Driver
    1.43
    als
    1.41
    ello
    1.40
     quotes
    1.40
    blogger
    1.39
    Activations Density 0.017%

    No Known Activations