© 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. 6839
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "ver."

    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
    45
    +0.13
    0.6%
    435
    +0.12
    0.6%
    727
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    435
    +0.13
    0.03
    545
    +0.12
    0.03
    284
    +0.12
    0.03
    Negative Logits
    ĪĴ
    -0.80
    erva
    -0.79
     millenn
    -0.79
    »Ĵ
    -0.70
    emonic
    -0.70
    encer
    -0.69
    oresc
    -0.68
    enced
    -0.65
    ĩ
    -0.65
     pestic
    -0.63
    POSITIVE LOGITS
    theless
    1.13
    ding
    0.95
    aging
    0.93
    deen
    0.91
    gence
    0.89
    aged
    0.87
    dict
    0.86
    tex
    0.82
    weight
    0.80
    ver
    0.79
    Activations Density 0.110%

    No Known Activations