© 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. 24591
    Prev
    Next
    INDEX
    Explanations

    phrases related to legal principles and outcomes

    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
    342
    +0.11
    0.6%
    98
    +0.10
    0.6%
    249
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    465
    +0.11
    0.15
    55
    +0.10
    0.07
    355
    +0.10
    0.12
    Negative Logits
    opes
    -1.45
    ires
    -1.44
    ĻĤ
    -1.40
    acial
    -1.37
     tons
    -1.37
    ork
    -1.33
    pretty
    -1.32
    locations
    -1.31
    ETHOD
    -1.29
     mornings
    -1.28
    POSITIVE LOGITS
    stown
    1.66
    Argued
    1.47
    :`
    1.47
    ressor
    1.45
    cheon
    1.45
    pora
    1.43
     indictment
    1.43
    issance
    1.42
    hood
    1.39
     referendum
    1.38
    Activations Density 3.289%

    No Known Activations