© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 30888
    Prev
    Next
    INDEX
    Explanations

    concepts related to optimization and improvement

    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
    111
    +0.11
    0.6%
    148
    +0.11
    0.6%
    376
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    267
    +0.11
    0.02
    134
    +0.11
    0.01
    326
    +0.11
    0.01
    Negative Logits
     slightest
    -1.82
    ?>
    -1.60
     ?>
    -1.58
     same
    -1.57
    ening
    -1.55
    mare
    -1.51
     superior
    -1.49
     presence
    -1.49
     respectively
    -1.43
    eners
    -1.41
    POSITIVE LOGITS
    ibus
    1.55
    imates
    1.50
    Ĩ
    1.47
    ais
    1.46
    ISPR
    1.44
    osomes
    1.43
    umns
    1.43
    in
    1.39
    packages
    1.38
    ista
    1.38
    Activations Density 0.075%

    No Known Activations