© 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. 27005
    Prev
    Next
    INDEX
    Explanations

    titles and positions related to leadership roles in legal 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
    226
    +0.13
    0.7%
    481
    +0.13
    0.7%
    254
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    254
    +0.13
    0.02
    298
    +0.13
    0.03
    501
    +0.11
    0.02
    Negative Logits
    ĻĤ
    -2.80
    »¿
    -2.41
    ¿½
    -2.31
    ´
    -2.15
    ¶
    -2.04
    ł
    -1.90
    ¾
    -1.88
    ½
    -1.79
    ª
    -1.71
    «
    -1.69
    POSITIVE LOGITS
    {{
    1.50
     {}
    1.49
     creators
    1.38
    Demo
    1.37
    Posts
    1.37
    Conclusion
    1.37
    template
    1.32
    achy
    1.32
    atos
    1.32
     divided
    1.31
    Activations Density 0.268%

    No Known Activations