© 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. 2-MLP-SM
    11. 24362
    Prev
    Next
    INDEX
    Explanations

    references to duality in various contexts

    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
    156
    +0.22
    1.3%
    183
    +0.14
    0.8%
    416
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    183
    +0.22
    0.02
    90
    +0.14
    0.02
    141
    +0.11
    0.03
    Negative Logits
    ¤
    -1.65
     Kong
    -1.32
     role
    -1.29
    ĺ
    -1.29
    ¶
    -1.27
     expressed
    -1.27
    bed
    -1.27
    ():
    -1.26
     ${{{\
    -1.25
    ½
    -1.25
    POSITIVE LOGITS
    ache
    1.61
    **).
    1.42
    axon
    1.37
     +=
    1.34
     |=
    1.33
    htra
    1.32
    itudes
    1.31
    visibility
    1.31
    rage
    1.30
    **),
    1.28
    Activations Density 0.272%

    No Known Activations