© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    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. 4775
    Prev
    Next
    INDEX
    Explanations

    terms associated with legal or regulatory concepts

    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
    373
    +0.13
    0.6%
    554
    +0.10
    0.4%
    215
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.13
    0.08
    284
    +0.10
    0.08
    373
    +0.10
    -0.04
    Negative Logits
     disability
    -0.63
     tongue
    -0.61
     block
    -0.61
    WHERE
    -0.60
    CLASS
    -0.60
     entitle
    -0.60
     spectrum
    -0.59
     coasts
    -0.59
     Hutch
    -0.59
     Scrolls
    -0.57
    POSITIVE LOGITS
    acci
    0.95
    etti
    0.94
    eros
    0.90
    arte
    0.89
    ulo
    0.89
    uez
    0.88
    ino
    0.87
    andise
    0.87
    anca
    0.84
    ierre
    0.80
    Activations Density 0.887%

    No Known Activations