© 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. 28348
    Prev
    Next
    INDEX
    Explanations

    words related to "us" and variations of the term "us."

    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
    445
    +0.12
    0.7%
    466
    +0.12
    0.7%
    271
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    445
    +0.12
    0.02
    271
    +0.12
    0.00
    343
    +0.11
    0.02
    Negative Logits
    your
    -1.59
     your
    -1.56
    typeof
    -1.53
    backs
    -1.45
     single
    -1.43
    bing
    -1.42
    middle
    -1.42
    ){#
    -1.39
    ){
    -1.34
     certain
    -1.32
    POSITIVE LOGITS
    aurus
    2.19
    pect
    2.07
    creen
    1.85
    criptor
    1.81
    cules
    1.79
    olve
    1.77
    cript
    1.77
    atsu
    1.75
    erve
    1.75
    burgh
    1.73
    Activations Density 0.115%

    No Known Activations