© 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. 28369
    Prev
    Next
    INDEX
    Explanations

    the repeated use of the word "the."

    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
    437
    +0.12
    0.7%
    235
    +0.11
    0.6%
    210
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    10
    +0.12
    0.01
    291
    +0.11
    0.01
    98
    +0.10
    0.00
    Negative Logits
    ables
    -1.57
    ering
    -1.52
     newsletters
    -1.52
     yourself
    -1.50
    icable
    -1.46
    lem
    -1.46
     any
    -1.45
    akers
    -1.45
     reductions
    -1.44
     rates
    -1.44
    POSITIVE LOGITS
    Decided
    1.97
    ·
    1.83
    áĢ¬
    1.71
    ³
    1.52
    ĻĤ
    1.45
    AndroidRuntime
    1.45
    ĥ½
    1.44
    áĢ¯
    1.43
    µ
    1.42
    conductor
    1.37
    Activations Density 0.001%

    No Known Activations