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

    Neuronpedia

    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. 0-MLP-SM
    11. 28294
    Prev
    Next
    INDEX
    Explanations

    instances of the word "IF" and related conditional language

    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
    +0.17
    1.0%
    231
    +0.10
    0.6%
    188
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    231
    +0.17
    0.01
    411
    +0.10
    0.01
    188
    +0.10
    0.01
    Negative Logits
    unreadable
    -1.82
    genesis
    -1.66
    activity
    -1.64
     disappear
    -1.52
    omitempty
    -1.42
     vanished
    -1.39
     childhood
    -1.39
     responded
    -1.38
     disappeared
    -1.35
     Nile
    -1.32
    POSITIVE LOGITS
    ière
    1.68
     scheme
    1.54
    CRA
    1.50
     aesthetics
    1.47
    reed
    1.46
    anguage
    1.46
    ung
    1.44
     version
    1.44
    ιÏĤ
    1.44
     alt
    1.40
    Activations Density 0.011%

    No Known Activations