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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    3. GPT2-Small
    4. Residual Post - 32k
    5. 8-RES_POST_32K-OAI
    6. 5081
    Prev
    Next
    INDEX
    Explanations

    statements of existence or location

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_8.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.8.hook_resid_post
    Hook Layer
    8
    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
    606
    +0.10
    0.4%
    314
    +0.09
    0.4%
    418
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    252
    +0.10
    0.02
    32
    +0.09
    0.02
    419
    +0.09
    0.02
    Negative Logits
    haw
    -0.72
    agers
    -0.66
    assian
    -0.64
    ivals
    -0.62
     rushes
    -0.62
    aires
    -0.61
    earances
    -0.60
    iles
    -0.60
     Barney
    -0.59
    oons
    -0.57
    POSITIVE LOGITS
     dormant
    1.11
     squarely
    0.98
    uten
    0.82
     somewhere
    0.80
     beneath
    0.80
     behind
    0.79
     Bene
    0.77
     rooted
    0.77
    Within
    0.76
     untouched
    0.76
    Activations Density 0.058%

    No Known Activations