© 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. 7581
    Prev
    Next
    INDEX
    Explanations

    phrases related to answering questions

    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
    452
    +0.11
    0.5%
    758
    +0.10
    0.5%
    719
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.11
    0.13
    608
    +0.10
    0.14
    163
    +0.10
    0.06
    Negative Logits
    acent
    -0.63
    raint
    -0.60
    ixie
    -0.60
    artifacts
    -0.58
    oult
    -0.58
    uren
    -0.57
    lite
    -0.57
    arbon
    -0.57
     bleach
    -0.57
    enta
    -0.56
    POSITIVE LOGITS
     questions
    1.62
     question
    1.39
     queries
    1.34
     inquiries
    1.28
     Questions
    1.18
     concerns
    1.18
     enqu
    1.18
     QUEST
    1.16
    quest
    1.15
    Question
    1.15
    Activations Density 4.830%

    No Known Activations