Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. Residuals
    5. 1-RES-SM
    6. 26812
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "Boost."

    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
    369
    +0.13
    0.8%
    385
    +0.13
    0.8%
    111
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    111
    +0.13
    0.01
    437
    +0.13
    0.01
    385
    +0.12
    0.01
    Negative Logits
    than
    -2.04
    upon
    -1.93
     touch
    -1.64
     groom
    -1.57
    Than
    -1.57
    however
    -1.51
    WHERE
    -1.50
    Plaintiff
    -1.50
     pleasure
    -1.48
    Defendants
    -1.47
    POSITIVE LOGITS
    ermann
    1.93
    forge
    1.74
    enix
    1.73
    ers
    1.73
    enos
    1.71
    emos
    1.71
    beans
    1.66
     liter
    1.64
    ños
    1.62
    etable
    1.61
    Activations Density 0.073%

    No Known Activations