© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 4945
    Prev
    Next
    INDEX
    Explanations

    references to interpersonal relationship status and legal 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
    459
    +0.16
    0.8%
    4
    +0.11
    0.5%
    285
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.16
    0.02
    404
    +0.11
    0.02
    4
    +0.10
    0.01
    Negative Logits
    é¾
    -0.70
    tower
    -0.70
    plin
    -0.68
    soDeliveryDate
    -0.66
    redits
    -0.66
    ĵĺ
    -0.66
    awaru
    -0.65
    isSpecialOrderable
    -0.63
    cheat
    -0.63
    ument
    -0.62
    POSITIVE LOGITS
    ized
    0.96
    ization
    0.95
    ity
    0.94
    isation
    0.91
    ogy
    0.91
    istical
    0.89
    isations
    0.89
    izations
    0.89
    icals
    0.88
    izing
    0.88
    Activations Density 0.072%

    No Known Activations