© 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. 2-MLP-SM
    11. 24283
    Prev
    Next
    INDEX
    Explanations

    quantitative values or numerical references in documents

    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/2-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.2.hook_mlp_out
    Hook Layer
    2
    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
    154
    +0.12
    0.7%
    451
    +0.12
    0.6%
    326
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    163
    +0.12
    0.01
    59
    +0.12
    0.01
    338
    +0.11
    0.02
    Negative Logits
    abit
    -2.13
    ystems
    -1.82
    imus
    -1.66
    mozilla
    -1.66
    documentclass
    -1.60
    enna
    -1.60
     here
    -1.49
    itos
    -1.49
    ATIONS
    -1.49
    izumab
    -1.46
    POSITIVE LOGITS
    ı
    4.16
    ®
    3.88
    ĻĤ
    3.84
    ¿½
    3.81
    ¾
    3.78
    ķ
    3.77
    Ĥ¬
    3.76
    ¦
    3.73
    IJ
    3.71
    ľ
    3.67
    Activations Density 0.209%

    No Known Activations