© 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
    3. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 8-RES_POST_32K-OAI
    11. 5392
    Prev
    Next
    INDEX
    Explanations

    terms related to law enforcement and public safety incidents

    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
    218
    +0.11
    0.5%
    431
    +0.11
    0.5%
    556
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.11
    0.20
    575
    +0.11
    0.16
    225
    +0.10
    0.12
    Negative Logits
    risome
    -0.76
    rieg
    -0.69
     attm
    -0.68
    é¾įå¥ij士
    -0.66
     externalToEVAOnly
    -0.66
    arth
    -0.66
    leck
    -0.66
    Lv
    -0.64
    yll
    -0.64
    Reloaded
    -0.63
    POSITIVE LOGITS
    reciation
    0.73
    jac
    0.69
     whom
    0.66
    iaries
    0.65
    iences
    0.64
     pronouns
    0.63
     Initiative
    0.62
     syndrome
    0.59
    itaire
    0.59
     waivers
    0.59
    Activations Density 2.730%

    No Known Activations