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. 4245
    Prev
    Next
    INDEX
    Explanations

    proper nouns and terms associated with governance or administrative entities

    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
    385
    +0.13
    0.6%
    23
    +0.12
    0.5%
    508
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    316
    +0.13
    0.02
    659
    +0.12
    0.01
    288
    +0.12
    0.02
    Negative Logits
    ees
    -0.69
    uate
    -0.68
    iliary
    -0.68
    illard
    -0.67
    ray
    -0.66
    =-=-=-=-
    -0.66
    ioned
    -0.64
    arded
    -0.64
    urst
    -0.63
     Thumbnails
    -0.63
    POSITIVE LOGITS
    conservancy
    0.93
    ktop
    0.93
    geist
    0.91
    baugh
    0.88
    ãĤº
    0.85
    ocide
    0.76
    cki
    0.75
    wald
    0.73
    artifacts
    0.73
    apons
    0.72
    Activations Density 0.086%

    No Known Activations