© 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. 24544
    Prev
    Next
    INDEX
    Explanations

    instances of refusal or denial related to legal or formal processes

    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
    192
    +0.13
    0.7%
    209
    +0.13
    0.7%
    416
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    5
    +0.13
    0.02
    485
    +0.13
    0.02
    314
    +0.13
    0.03
    Negative Logits
     opponent
    -1.66
     opponents
    -1.62
     extremes
    -1.56
     defence
    -1.56
     limitation
    -1.55
     orders
    -1.54
     viz
    -1.53
     overex
    -1.52
     predecessor
    -1.52
     measuring
    -1.51
    POSITIVE LOGITS
     enroll
    1.64
     await
    1.63
    _."
    1.59
    izon
    1.57
    ?_
    1.57
    ierno
    1.54
    Īĺ
    1.51
    ĭ
    1.49
    inux
    1.49
    izumab
    1.47
    Activations Density 1.437%

    No Known Activations