© 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. 8071
    Prev
    Next
    INDEX
    Explanations

    elements specific to research methodology and academic terminology

    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
    373
    +0.13
    0.6%
    339
    +0.09
    0.4%
    607
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    696
    +0.13
    0.05
    739
    +0.09
    0.02
    167
    +0.08
    -0.00
    Negative Logits
    llular
    -0.62
     HOU
    -0.62
    dB
    -0.61
    FY
    -0.61
    daq
    -0.60
     Geo
    -0.59
     Nanto
    -0.58
    ischer
    -0.58
     DOT
    -0.58
    outer
    -0.58
    POSITIVE LOGITS
    ccording
    1.02
    itionally
    0.72
    ritis
    0.70
    neys
    0.69
    ©¶æ
    0.69
    oubtedly
    0.69
    ometime
    0.68
    onga
    0.67
    ¶
    0.67
    VERTISEMENT
    0.65
    Activations Density 0.780%

    No Known Activations