© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 10-RES_POST_32K-OAI
    11. 28374
    Prev
    Next
    INDEX
    Explanations

    references to leadership or guidance in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_10.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    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
    319
    +0.10
    0.4%
    361
    +0.10
    0.4%
    269
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    443
    +0.10
    0.02
    225
    +0.10
    0.04
    646
    +0.10
    0.03
    Negative Logits
     contrace
    -0.64
    ãĥĩãĤ£
    -0.63
    IOR
    -0.63
    itary
    -0.63
    roxy
    -0.62
    Area
    -0.61
    tu
    -0.59
    pora
    -0.59
    afety
    -0.58
    oln
    -0.58
    POSITIVE LOGITS
     footsteps
    1.53
     directions
    1.17
     path
    1.14
     instructions
    1.09
     lead
    1.06
     footprints
    1.03
     trail
    1.02
     paths
    1.02
     route
    0.98
     trajectory
    0.97
    Activations Density 1.290%

    No Known Activations