© 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. 8525
    Prev
    Next
    INDEX
    Explanations

    references to "the behind" or "behind the scenes" contexts

    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
    572
    +0.10
    0.5%
    520
    +0.09
    0.4%
    691
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.10
    0.04
    476
    +0.09
    0.02
    572
    +0.09
    0.02
    Negative Logits
    RIC
    -0.67
    arily
    -0.67
     Heist
    -0.63
     msec
    -0.59
    ERG
    -0.58
    odder
    -0.58
     breeze
    -0.57
    aez
    -0.57
    pring
    -0.57
    ital
    -0.56
    POSITIVE LOGITS
     scenes
    1.26
     curtain
    1.02
     Scenes
    1.01
    scenes
    0.89
     curtains
    0.80
     curve
    0.73
     veil
    0.67
     lines
    0.67
     wheel
    0.66
     mask
    0.65
    Activations Density 0.251%

    No Known Activations