© 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. 2-MLP-SM
    11. 25707
    Prev
    Next
    INDEX
    Explanations

    statements related to governmental issues and public health concerns

    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
    257
    +0.14
    0.8%
    455
    +0.13
    0.7%
    508
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    508
    +0.14
    0.09
    182
    +0.13
    0.11
    274
    +0.12
    0.09
    Negative Logits
    ĻĤ
    -3.19
    ¿½
    -2.18
    ĨĴ
    -2.14
    ´
    -2.13
    ³
    -2.10
    Ļª
    -2.08
    ĵ
    -2.05
    ª
    -2.04
    »¿
    -1.97
    ĭ
    -1.93
    POSITIVE LOGITS
    iei
    1.74
    plots
    1.64
    <_
    1.54
    Davis
    1.52
    prints
    1.52
    lights
    1.51
    Plot
    1.51
    lessly
    1.48
    ably
    1.48
     comprising
    1.47
    Activations Density 2.139%

    No Known Activations