© 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. 29998
    Prev
    Next
    INDEX
    Explanations

    words and phrases associated with death and dying

    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
    239
    +0.16
    0.9%
    376
    +0.12
    0.7%
    46
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    239
    +0.16
    0.02
    46
    +0.12
    0.01
    505
    +0.12
    0.01
    Negative Logits
    inberg
    -1.71
    aded
    -1.69
    ...](
    -1.58
    ores
    -1.51
    quired
    -1.48
    ONS
    -1.48
    hern
    -1.47
    rolling
    -1.45
    anc
    -1.41
    ]]
    -1.41
    POSITIVE LOGITS
    liest
    1.74
     beforehand
    1.69
    ––
    1.52
     bases
    1.48
    cule
    1.47
    bolt
    1.47
    ?,
    1.47
     before
    1.44
     tipo
    1.42
     cycle
    1.42
    Activations Density 0.021%

    No Known Activations