© 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. 25765
    Prev
    Next
    INDEX
    Explanations

    legal terms and references in court cases

    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
    +0.14
    0.8%
    275
    +0.14
    0.8%
    207
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    135
    +0.14
    0.03
    44
    +0.14
    0.01
    348
    +0.11
    0.01
    Negative Logits
     NEGLIGENCE
    -1.53
    )|$(
    -1.53
     WARRANTY
    -1.50
    icul
    -1.46
     confusion
    -1.41
    ahoma
    -1.39
     progress
    -1.38
     antitrust
    -1.35
     privacy
    -1.34
     polit
    -1.34
    POSITIVE LOGITS
    ĩ
    2.72
    ¼
    2.69
    ¢
    2.60
    ĻĤ
    2.53
    ij
    2.48
    Ŀ
    2.42
    ĥ½
    2.40
    £
    2.40
    ·
    2.31
    Ĭ
    2.31
    Activations Density 0.305%

    No Known Activations