© 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. 30484
    Prev
    Next
    INDEX
    Explanations

    instances of the word "only."

    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
    749
    +0.10
    0.5%
    183
    +0.10
    0.4%
    28
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    512
    +0.10
    0.02
    413
    +0.10
    0.02
    184
    +0.09
    0.01
    Negative Logits
    »Ĵ
    -0.93
    EStream
    -0.73
    etc
    -0.70
    ascript
    -0.66
    ĺħ
    -0.63
    ij士
    -0.63
    Enlarge
    -0.62
    ashington
    -0.61
    yip
    -0.61
    ahime
    -0.61
    POSITIVE LOGITS
    oped
    0.76
    ifiable
    0.71
     physically
    0.70
     tolerated
    0.68
     financially
    0.63
    iani
    0.63
    astically
    0.62
    hetical
    0.62
     visually
    0.61
     survive
    0.60
    Activations Density 0.392%

    No Known Activations