© 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. Residuals
    9. 1-RES-SM
    11. 25335
    Prev
    Next
    INDEX
    Explanations

    mathematical expressions and symbols, particularly those related to equations or inequalities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__res-sm_processed/1-res-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.1.hook_resid_post
    Hook Layer
    1
    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
    188
    +0.18
    1.1%
    219
    +0.17
    1.0%
    478
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    188
    +0.18
    0.08
    219
    +0.17
    0.08
    380
    +0.14
    0.05
    Negative Logits
    omed
    -1.89
    anged
    -1.67
    oused
    -1.58
    oved
    -1.56
    amon
    -1.50
    Than
    -1.48
     Briefly
    -1.47
    eric
    -1.46
    anging
    -1.45
    pled
    -1.44
    POSITIVE LOGITS
    ĻĤ
    2.00
    İ
    1.99
    Ī
    1.93
    ĥ
    1.91
    «
    1.85
    ´
    1.84
    º
    1.82
    ı
    1.77
    Īĺ
    1.74
    ĺ
    1.73
    Activations Density 0.265%

    No Known Activations