© 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. 8-RES_POST_32K-OAI
    11. 6574
    Prev
    Next
    INDEX
    Explanations

    file types and configuration settings related to programming and scripting

    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
    524
    +0.15
    0.7%
    545
    +0.11
    0.5%
    517
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.15
    0.04
    158
    +0.11
    0.04
    542
    +0.10
    0.04
    Negative Logits
    ãĤ¼
    -0.73
    inav
    -0.71
    abouts
    -0.70
    bent
    -0.68
     Orchestra
    -0.62
     PACs
    -0.61
    itism
    -0.61
     Samar
    -0.60
     pals
    -0.60
    opoulos
    -0.60
    POSITIVE LOGITS
     file
    1.28
     FILE
    1.08
     files
    1.03
     File
    0.94
    tml
    0.91
     Files
    0.85
     format
    0.85
     extension
    0.84
    file
    0.84
     template
    0.82
    Activations Density 0.291%

    No Known Activations