© 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. 28908
    Prev
    Next
    INDEX
    Explanations

    descriptions and attributes of physical spaces or objects

    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
    547
    +0.12
    0.5%
    312
    +0.10
    0.5%
    598
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    678
    +0.12
    0.07
    289
    +0.10
    0.00
    362
    +0.10
    0.08
    Negative Logits
     conversions
    -0.78
    topic
    -0.72
    RESULTS
    -0.72
     destinations
    -0.71
     outcomes
    -0.70
     Topics
    -0.67
    imeo
    -0.67
     Rates
    -0.66
     Views
    -0.65
    pring
    -0.64
    POSITIVE LOGITS
     veil
    1.33
     shroud
    1.23
     curtains
    1.19
     protective
    1.18
     curtain
    1.15
     secrecy
    1.10
     cloak
    1.06
     blanket
    1.04
     blankets
    1.03
     shielding
    1.03
    Activations Density 2.054%

    No Known Activations