© 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. 0-MLP-SM
    11. 28729
    Prev
    Next
    INDEX
    Explanations

    the term "house" in various contexts, particularly in descriptions of properties or structural features

    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
    118
    +0.12
    0.7%
    347
    +0.11
    0.7%
    191
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    347
    +0.12
    0.02
    34
    +0.11
    0.02
    388
    +0.11
    0.02
    Negative Logits
    '?"
    -1.86
    "?"
    -1.86
     thee
    -1.74
     snow
    -1.73
    dle
    -1.66
     Parenthood
    -1.66
     APPEALS
    -1.63
    .'"
    -1.61
    ?'
    -1.61
    ?’
    -1.61
    POSITIVE LOGITS
    ware
    2.40
    hold
    2.03
    holds
    2.02
    plots
    1.93
    holder
    1.88
    steps
    1.75
    notes
    1.73
    cule
    1.73
    ings
    1.70
    keeping
    1.70
    Activations Density 0.016%

    No Known Activations