© 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. 28412
    Prev
    Next
    INDEX
    Explanations

    references to the concept of "behind the scenes" or hidden aspects

    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
    520
    +0.12
    0.5%
    572
    +0.11
    0.5%
    454
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.12
    0.05
    374
    +0.11
    -0.00
    526
    +0.10
    -0.00
    Negative Logits
    cision
    -0.78
    RIC
    -0.77
    ultimate
    -0.68
    ital
    -0.63
     secondly
    -0.62
    ryu
    -0.60
    ommod
    -0.59
    ornia
    -0.58
    borg
    -0.58
    iple
    -0.57
    POSITIVE LOGITS
     scenes
    1.56
    scenes
    1.24
     Scenes
    1.19
     curtain
    1.10
     wheel
    0.95
     curtains
    0.92
    scene
    0.79
     desk
    0.78
     curve
    0.77
     facade
    0.75
    Activations Density 0.408%

    No Known Activations