© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. MLP Post
    9. 0-MLP-SM
    11. 30604
    Prev
    Next
    INDEX
    Explanations

    instances of the word "disclose" related to transparency and revelation of information

    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/0-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.0.hook_mlp_out
    Hook Layer
    0
    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
    376
    +0.25
    1.4%
    148
    +0.12
    0.7%
    413
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    167
    +0.25
    0.01
    74
    +0.12
    0.01
    428
    +0.11
    0.01
    Negative Logits
    uten
    -1.88
     sued
    -1.51
    ţi
    -1.50
    agner
    -1.42
    ż
    -1.41
    ‘
    -1.41
    ived
    -1.38
     «
    -1.36
    «
    -1.35
    Äį
    -1.33
    POSITIVE LOGITS
    ĥ
    1.61
    ª
    1.59
    ence
    1.56
    cibly
    1.44
     nonsense
    1.42
    yt
    1.42
    ulate
    1.41
    °
    1.38
    olate
    1.37
    mouth
    1.34
    Activations Density 0.284%

    No Known Activations