© 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. 4820
    Prev
    Next
    INDEX
    Explanations

    mentions of the word "at" in various contexts

    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
    61
    +0.11
    0.5%
    724
    +0.11
    0.5%
    499
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    371
    +0.11
    0.05
    362
    +0.11
    0.05
    62
    +0.10
    0.04
    Negative Logits
     arous
    -0.73
    ¥ŀ
    -0.73
    omething
    -0.70
    clusively
    -0.70
     settling
    -0.69
     withdrawing
    -0.66
     depos
    -0.65
     swapping
    -0.64
    oké
    -0.64
    avorite
    -0.63
    POSITIVE LOGITS
    rix
    0.72
    =-=-=-=-
    0.70
    roit
    0.68
    ols
    0.67
    yy
    0.67
    ragon
    0.67
     Large
    0.67
    v
    0.66
    ariat
    0.66
    rod
    0.66
    Activations Density 0.324%

    No Known Activations