© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. MLP Post
    9. 2-MLP-SM
    11. 23721
    Prev
    Next
    INDEX
    Explanations

    references to legal terminology and concepts 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/2-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.2.hook_mlp_out
    Hook Layer
    2
    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
    397
    +0.15
    0.8%
    312
    +0.13
    0.7%
    439
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    250
    +0.15
    0.03
    397
    +0.13
    0.04
    204
    +0.12
    0.01
    Negative Logits
     asylum
    -1.61
     rows
    -1.43
    ISTS
    -1.42
    ringes
    -1.41
    hetized
    -1.39
    WO
    -1.39
    awed
    -1.37
     Grove
    -1.36
     mock
    -1.33
     REST
    -1.32
    POSITIVE LOGITS
    áĢ¬
    1.80
    áĢIJ
    1.75
    àµį
    1.57
     doi
    1.57
    áĢĢ
    1.45
     dise
    1.38
    ariat
    1.38
    eron
    1.37
    áĢ¯
    1.36
    ¥
    1.35
    Activations Density 0.385%

    No Known Activations