© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 28443
    Prev
    Next
    INDEX
    Explanations

    formal legal language and terminology

    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
    342
    +0.19
    1.1%
    286
    +0.16
    0.9%
    320
    +0.15
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    342
    +0.19
    0.07
    316
    +0.16
    0.07
    55
    +0.15
    0.09
    Negative Logits
    ));
    -1.71
    owe
    -1.69
     donkey
    -1.44
    woke
    -1.43
     respectively
    -1.38
    iles
    -1.34
    Īĺ
    -1.32
    acts
    -1.32
    amid
    -1.31
     theirs
    -1.31
    POSITIVE LOGITS
    strom
    1.70
    itness
    1.63
    iab
    1.50
    istar
    1.38
     unfolded
    1.36
    omen
    1.35
    dorff
    1.33
    IEW
    1.27
    iele
    1.26
     herself
    1.25
    Activations Density 3.366%

    No Known Activations