© 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. 29355
    Prev
    Next
    INDEX
    Explanations

    terms related to reversal or overturning decisions and situations

    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
    376
    +0.15
    0.9%
    371
    +0.12
    0.7%
    156
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    371
    +0.15
    0.01
    212
    +0.12
    0.01
    178
    +0.11
    0.01
    Negative Logits
    FFIR
    -1.71
    EVER
    -1.67
    >'
    -1.66
    obbsee
    -1.64
    usions
    -1.61
    mail
    -1.59
    &=&\
    -1.58
    aiver
    -1.58
    bey
    -1.58
    >';
    -1.57
    POSITIVE LOGITS
    ists
    1.66
    ist
    1.60
     chambers
    1.60
     settings
    1.58
    ÅĽci
    1.58
     positions
    1.57
    arily
    1.54
     groups
    1.51
     order
    1.51
     compartments
    1.49
    Activations Density 0.017%

    No Known Activations