© 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. 28124
    Prev
    Next
    INDEX
    Explanations

    repeated mentions of the word "consent."

    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
    +0.15
    0.9%
    83
    +0.13
    0.7%
    112
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    83
    +0.15
    0.03
    112
    +0.13
    0.02
    194
    +0.12
    0.02
    Negative Logits
    ¯
    -2.64
    ±
    -2.58
    ¿
    -2.56
    ĨĴ
    -2.55
    ¦
    -2.51
    ĻĤ
    -2.50
    Ĺ
    -2.50
    Ł
    -2.44
    ĩ
    -2.44
    ©
    -2.42
    POSITIVE LOGITS
    onant
    1.90
    cript
    1.72
    omination
    1.58
    umption
    1.54
    piring
    1.51
    imony
    1.51
    pect
    1.51
    itution
    1.50
    olation
    1.50
    igator
    1.46
    Activations Density 0.228%

    No Known Activations