© 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. 30106
    Prev
    Next
    INDEX
    Explanations

    phrases denoting physical locations and barriers

    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
    464
    +0.10
    0.4%
    745
    +0.08
    0.4%
    125
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.10
    0.12
    249
    +0.08
    0.04
    16
    +0.08
    0.03
    Negative Logits
    ombies
    -0.78
     Typh
    -0.76
    astical
    -0.71
    oneliness
    -0.68
    Week
    -0.66
    folios
    -0.66
     Chronicles
    -0.66
    ģĸ
    -0.66
    humans
    -0.66
    itic
    -0.65
    POSITIVE LOGITS
     driveway
    1.44
     alley
    1.44
     doorway
    1.44
     entrance
    1.41
     sidewalk
    1.38
     gate
    1.35
     door
    1.34
     intersection
    1.33
     porch
    1.28
     exit
    1.27
    Activations Density 1.447%

    No Known Activations