© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 0-MLP-SM
    11. 26775
    Prev
    Next
    INDEX
    Explanations

    the word "allegedly."

    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
    376
    +0.20
    1.2%
    156
    +0.20
    1.2%
    148
    +0.15
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    204
    +0.20
    0.02
    322
    +0.20
    0.02
    167
    +0.15
    0.02
    Negative Logits
    Ī
    -3.05
    Ĩ
    -2.67
    ĨĴ
    -2.59
    Ģ
    -2.58
    ĭ
    -2.54
    ķ
    -2.51
    ↵↵
    -2.49
    -2.49
    -2.49
    č↵
    -2.49
    POSITIVE LOGITS
     ration
    1.54
     hers
    1.44
     Murder
    1.44
     quar
    1.40
    itiveness
    1.37
     punish
    1.35
     necrosis
    1.34
    evil
    1.33
     contradictory
    1.33
    abol
    1.31
    Activations Density 0.012%

    No Known Activations