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. 2-RES-SM
    6. 29421
    Prev
    Next
    INDEX
    Explanations

    data related to population statistics and demographic information

    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
    63
    +0.12
    0.7%
    313
    +0.11
    0.6%
    204
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    385
    +0.12
    0.01
    288
    +0.11
    0.02
    497
    +0.11
    0.01
    Negative Logits
    lectric
    -1.50
    mathscr
    -1.45
    while
    -1.45
    "}](#
    -1.44
    router
    -1.43
    act
    -1.43
    suits
    -1.38
    cleanup
    -1.37
    ellees
    -1.37
    '))
    -1.31
    POSITIVE LOGITS
     age
    1.58
     Caption
    1.56
     Lik
    1.53
     popularity
    1.48
     ratings
    1.48
    —.
    1.44
     veteran
    1.41
    adian
    1.41
    thood
    1.39
     accession
    1.33
    Activations Density 0.164%

    No Known Activations