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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    3. GPT2-Small
    4. Residual Post - 32k
    5. 6-RES_POST_32K-OAI
    6. 4213
    Prev
    Next
    INDEX
    Explanations

    articles and quantifiers related to singular or plural nouns

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 6-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_6.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.6.hook_resid_post
    Hook Layer
    6
    Architecture
    standard
    Context Size
    64
    Dataset
    Skylion007/openwebtext
    Activation Function
    topk
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    421
    +0.10
    0.5%
    310
    +0.10
    0.5%
    287
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    36
    +0.10
    0.06
    393
    +0.10
    0.04
    310
    +0.10
    0.03
    Negative Logits
    kefeller
    -0.87
    jri
    -0.74
    aptic
    -0.71
    RFC
    -0.70
    thumbnails
    -0.70
    achine
    -0.69
    rake
    -0.68
    apult
    -0.68
    erning
    -0.66
    gression
    -0.64
    POSITIVE LOGITS
     lucky
    0.82
     fortunate
    0.79
     kidding
    0.71
     Quite
    0.68
     aver
    0.66
    oooo
    0.65
     Nig
    0.63
     Few
    0.63
     Investigator
    0.62
     brill
    0.61
    Activations Density 0.228%

    No Known Activations