© 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. 5300
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "two" and variations of the number two in different 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
    553
    +0.10
    0.4%
    314
    +0.09
    0.4%
    694
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    693
    +0.10
    0.04
    694
    +0.09
    0.02
    183
    +0.09
    0.03
    Negative Logits
     guiName
    -0.76
     guy
    -0.70
     itself
    -0.68
     Skydragon
    -0.67
     quantity
    -0.65
    maxwell
    -0.65
    icter
    -0.64
    cair
    -0.62
     outweigh
    -0.62
     thing
    -0.62
    POSITIVE LOGITS
     own
    1.16
     favorite
    1.08
    favorite
    0.97
     favourite
    0.93
     predecessors
    0.93
     Own
    0.92
     Favorite
    0.87
     foray
    0.80
     cents
    0.77
    own
    0.76
    Activations Density 0.264%

    No Known Activations