© 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. 4935
    Prev
    Next
    INDEX
    Explanations

    terms related to specific types of media, particularly those used in fine arts or entertainment

    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
    699
    +0.10
    0.5%
    89
    +0.10
    0.4%
    71
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    701
    +0.10
    0.02
    406
    +0.10
    0.02
    659
    +0.10
    0.02
    Negative Logits
    enegger
    -0.87
    ãĥ¤
    -0.62
    door
    -0.60
     resil
    -0.59
     doors
    -0.58
     Aires
    -0.57
    ãĥ¼ãĥĨ
    -0.56
    nces
    -0.56
     ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³
    -0.56
    riott
    -0.56
    POSITIVE LOGITS
    nesota
    1.00
    ota
    0.83
    OTA
    0.81
    ippi
    0.81
    amps
    0.70
    iae
    0.68
    orum
    0.66
    kees
    0.63
    oes
    0.63
    igure
    0.63
    Activations Density 0.163%

    No Known Activations