© 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. 29314
    Prev
    Next
    INDEX
    Explanations

    punctuation marks that indicate pauses or separations in text

    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
    195
    +0.12
    0.7%
    311
    +0.11
    0.6%
    435
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    203
    +0.12
    0.36
    23
    +0.11
    0.31
    494
    +0.10
    0.25
    Negative Logits
    eno
    -1.84
    enos
    -1.63
     год
    -1.55
    fired
    -1.54
    zeit
    -1.52
    cats
    -1.51
    ned
    -1.45
    erne
    -1.45
    etime
    -1.44
    AZ
    -1.44
    POSITIVE LOGITS
     nowhere
    1.58
     places
    1.56
    ³
    1.51
     shoulders
    1.50
    ĥ
    1.48
     hearts
    1.47
    Ĺ
    1.41
     worlds
    1.41
     faults
    1.39
     lives
    1.36
    Activations Density 3.067%

    No Known Activations