© 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. 2-RES-SM
    11. 31355
    Prev
    Next
    INDEX
    Explanations

    mathematical expressions involving symbols and braces

    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/2-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.2.hook_resid_post
    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
    478
    +0.16
    0.9%
    320
    +0.13
    0.7%
    4
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    460
    +0.16
    0.01
    175
    +0.13
    0.01
    487
    +0.12
    0.01
    Negative Logits
    instein
    -1.66
    documentclass
    -1.63
    acted
    -1.62
    emed
    -1.55
    ctions
    -1.55
    ↵³³
    -1.53
    through
    -1.48
    Than
    -1.47
    mog
    -1.45
    msgstr
    -1.45
    POSITIVE LOGITS
    nis
    1.57
     doping
    1.53
     fed
    1.44
    ĥ½
    1.44
    leftrightarrow
    1.40
     biodegrad
    1.36
     slightest
    1.36
     infinites
    1.32
    acity
    1.31
     synthetic
    1.30
    Activations Density 0.052%

    No Known Activations