© 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 EvalsExportsinterp-engineNEWAPI 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. 7473
    Prev
    Next
    INDEX
    Explanations

    patterns related to classification or categorization of data

    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
    716
    +0.10
    0.5%
    331
    +0.09
    0.4%
    55
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    627
    +0.10
    0.03
    373
    +0.09
    -0.03
    316
    +0.09
    0.04
    Negative Logits
     Towers
    -0.82
     Salam
    -0.82
     targ
    -0.80
     Tags
    -0.78
     stacking
    -0.74
    Tile
    -0.74
     OT
    -0.73
     tile
    -0.73
     Rag
    -0.72
     TA
    -0.71
    POSITIVE LOGITS
    19
    1.10
    191
    1.07
     191
    1.06
    190
    1.00
     193
    1.00
    193
    0.99
    194
    0.99
     19
    0.98
     190
    0.98
     189
    0.98
    Activations Density 0.753%

    No Known Activations