© 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. 28224
    Prev
    Next
    INDEX
    Explanations

    references to specific written notes or documentation

    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
    +0.19
    1.1%
    273
    +0.16
    0.9%
    100
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    69
    +0.19
    0.03
    109
    +0.16
    0.03
    434
    +0.13
    0.02
    Negative Logits
    ĥ½
    -2.84
    ¼
    -2.45
    ı
    -2.33
    Ī
    -2.25
    ¸
    -2.09
    -2.08
    -2.08
    ↵
    -2.08
       ↵
    -2.08
    <|outofrange|>
    -2.08
    POSITIVE LOGITS
    ég
    1.84
    cule
    1.81
    Bank
    1.79
    iano
    1.77
    CoV
    1.76
    erve
    1.69
    naire
    1.66
    icile
    1.60
    ylvan
    1.59
    ulent
    1.59
    Activations Density 1.685%

    No Known Activations