© 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. 28594
    Prev
    Next
    INDEX
    Explanations

    various forms of the verb "to be."

    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
    197
    +0.13
    0.7%
    466
    +0.10
    0.6%
    339
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    509
    +0.13
    0.05
    339
    +0.10
    0.03
    380
    +0.10
    0.04
    Negative Logits
    ĵ
    -2.79
    ĥ½
    -2.69
    Į
    -2.42
    ķ
    -2.40
    ĺ
    -2.27
    ļ
    -2.23
    ↵↵
    -2.23
    -2.23
    č↵č↵
    -2.23
    ↵
    -2.23
    POSITIVE LOGITS
    ings
    1.67
    buck
    1.57
    PING
    1.56
    estone
    1.50
    croft
    1.50
     sides
    1.49
    doms
    1.44
    quel
    1.44
    ledge
    1.44
    ellaneous
    1.42
    Activations Density 0.480%

    No Known Activations