© 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. 24495
    Prev
    Next
    INDEX
    Explanations

    terms related to antidepressants or related medications

    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
    322
    +0.11
    0.6%
    410
    +0.11
    0.6%
    158
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    410
    +0.11
    0.05
    457
    +0.11
    0.03
    158
    +0.10
    0.08
    Negative Logits
    ¸
    -1.98
    ķ
    -1.74
    ľ
    -1.74
    lop
    -1.70
    weg
    -1.58
    İ
    -1.55
    card
    -1.54
    itarian
    -1.53
    rak
    -1.52
    ard
    -1.51
    POSITIVE LOGITS
     arranged
    1.57
     end
    1.52
     Esq
    1.50
    minster
    1.41
     representatives
    1.41
     embargo
    1.39
     acting
    1.39
     UAE
    1.34
     deepest
    1.31
     eager
    1.31
    Activations Density 1.373%

    No Known Activations