© 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. 30930
    Prev
    Next
    INDEX
    Explanations

    phrases or concepts associated with significant quantities or foundational aspects within a context

    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
    317
    +0.14
    0.8%
    413
    +0.13
    0.7%
    209
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    494
    +0.14
    0.36
    402
    +0.13
    0.21
    492
    +0.12
    0.21
    Negative Logits
     Papers
    -1.71
    «
    -1.53
     heritage
    -1.51
    ogly
    -1.40
    manual
    -1.36
    UTERS
    -1.35
    itative
    -1.33
     fat
    -1.32
     alive
    -1.32
     markup
    -1.29
    POSITIVE LOGITS
    **](#
    1.46
    achus
    1.44
     Idaho
    1.41
    rangle
    1.39
    á¿ĸ
    1.36
    anting
    1.34
    7554
    1.33
    venth
    1.33
    ref
    1.32
     instant
    1.32
    Activations Density 2.687%

    No Known Activations