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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 26428
    Prev
    Next
    INDEX
    Explanations

    specific legal or regulatory terms and phrases

    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
    23
    +0.11
    0.6%
    268
    +0.11
    0.6%
    461
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    120
    +0.11
    0.03
    189
    +0.11
    0.02
    343
    +0.10
    0.02
    Negative Logits
    sworth
    -1.62
    tz
    -1.52
     ("
    -1.46
     (@
    -1.46
     Weekly
    -1.44
    cias
    -1.44
     email
    -1.43
     into
    -1.43
     Publishers
    -1.42
     called
    -1.42
    POSITIVE LOGITS
    ĸ
    2.10
    »¿
    2.00
    ¦
    1.95
    ĸ´
    1.91
    Ł
    1.91
    ¼
    1.91
    ı
    1.89
    ¤
    1.89
    ļ
    1.88
    »
    1.82
    Activations Density 0.220%

    No Known Activations