© 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. 5409
    Prev
    Next
    INDEX
    Explanations

    the presence of geographic or location-related terms

    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
    +0.11
    0.5%
    530
    +0.09
    0.4%
    584
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.11
    0.05
    232
    +0.09
    0.04
    284
    +0.09
    0.03
    Negative Logits
    irlf
    -0.80
    Ħ¢
    -0.65
     anonymously
    -0.57
    imental
    -0.57
     Tide
    -0.56
     constitu
    -0.56
     Hatch
    -0.55
     prob
    -0.54
    aucus
    -0.54
    lege
    -0.54
    POSITIVE LOGITS
     Reloaded
    0.65
    hard
    0.63
    inka
    0.61
    gow
    0.61
    cules
    0.60
    ritic
    0.60
    andra
    0.59
    urgy
    0.59
     Rooms
    0.59
    anie
    0.58
    Activations Density 0.260%

    No Known Activations