© 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. 27826
    Prev
    Next
    INDEX
    Explanations

    multiple references to academic or official documents and reports

    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
    156
    +0.24
    1.4%
    287
    +0.11
    0.7%
    376
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    287
    +0.24
    0.01
    315
    +0.11
    0.01
    198
    +0.10
    0.01
    Negative Logits
    ĥ½
    -3.34
    ı
    -2.64
    §
    -2.63
    ↵
    -2.37
    ↵↵
    -2.37
    -2.37
    <|outofrange|>
    -2.37
    -2.37
    -2.37
    -2.37
    POSITIVE LOGITS
    matter
    1.71
    sense
    1.71
    iary
    1.60
    area
    1.58
    flash
    1.57
    ious
    1.50
    lot
    1.49
    mortem
    1.46
    plate
    1.42
    boarding
    1.42
    Activations Density 0.016%

    No Known Activations