© 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. 29846
    Prev
    Next
    INDEX
    Explanations

    references to empirical evidence or data

    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
    +0.39
    2.4%
    376
    +0.17
    1.0%
    407
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    156
    +0.39
    0.01
    320
    +0.17
    0.01
    407
    +0.13
    0.01
    Negative Logits
    Į
    -2.13
    ĥ½
    -1.87
    §
    -1.75
    ĵ
    -1.61
    ī
    -1.58
    ©
    -1.50
    ĨĴ
    -1.46
    ĩ
    -1.45
    ¯
    -1.44
    anov
    -1.44
    POSITIVE LOGITS
    het
    1.51
     insertions
    1.47
    guide
    1.46
    idades
    1.43
    hors
    1.42
    fiddle
    1.41
    iper
    1.39
    positions
    1.37
    find
    1.36
    scripts
    1.33
    Activations Density 0.013%

    No Known Activations