© 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. 28492
    Prev
    Next
    INDEX
    Explanations

    instances of opening parentheses in code or programming contexts

    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
    457
    +0.13
    0.7%
    478
    +0.12
    0.7%
    129
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    308
    +0.13
    0.01
    129
    +0.12
    0.01
    112
    +0.11
    0.01
    Negative Logits
    ians
    -2.01
    iano
    -1.97
    ena
    -1.92
    ian
    -1.84
    \]](
    -1.65
    smith
    -1.64
    onom
    -1.57
     Mell
    -1.54
    iah
    -1.52
    olver
    -1.46
    POSITIVE LOGITS
    ĻĤ
    1.58
     veins
    1.46
     distances
    1.45
    ?)
    1.43
    ...)
    1.41
    |}
    1.38
    athe
    1.35
    ľ
    1.34
    however
    1.32
    ĥ½
    1.30
    Activations Density 0.006%

    No Known Activations