© 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. 28676
    Prev
    Next
    INDEX
    Explanations

    mentions of public relations or PR terminology

    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
    369
    +0.15
    0.9%
    376
    +0.12
    0.7%
    285
    +0.11
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    67
    +0.15
    0.02
    285
    +0.12
    0.02
    480
    +0.11
    0.02
    Negative Logits
    Ļª
    -2.44
    ¼
    -2.27
    ľĵ
    -2.18
    ¿
    -1.88
    ij
    -1.81
    ĥ½
    -1.80
    hood
    -1.79
    Ĺ
    -1.79
    ķ
    -1.77
    ¢
    -1.75
    POSITIVE LOGITS
     genre
    1.81
    ODUCT
    1.80
     conventions
    1.73
     stunt
    1.72
    ODUCTION
    1.71
    IMARY
    1.69
     legend
    1.66
    ings
    1.64
    film
    1.63
    ifer
    1.62
    Activations Density 0.005%

    No Known Activations