© 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. 4807
    Prev
    Next
    INDEX
    Explanations

    references to vampires and horror-related themes

    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
    346
    +0.10
    0.5%
    719
    +0.10
    0.5%
    375
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.10
    0.05
    326
    +0.10
    0.04
    151
    +0.09
    0.04
    Negative Logits
    cially
    -0.83
    estamp
    -0.80
    itsch
    -0.77
    icion
    -0.73
    çīĪ
    -0.72
    erate
    -0.71
    uments
    -0.70
    lee
    -0.70
    eday
    -0.69
    ports
    -0.69
    POSITIVE LOGITS
     craw
    0.95
     Bane
    0.92
     lord
    0.88
     Dracula
    0.85
     lords
    0.82
     Masquerade
    0.81
     plague
    0.81
     crawl
    0.79
    evil
    0.78
     incarn
    0.77
    Activations Density 0.189%

    No Known Activations