© 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. 6625
    Prev
    Next
    INDEX
    Explanations

    conversations or statements related to awareness and responsiveness regarding specific topics or issues

    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
    604
    +0.11
    0.5%
    219
    +0.10
    0.5%
    416
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    196
    +0.11
    0.10
    661
    +0.10
    0.07
    685
    +0.10
    0.05
    Negative Logits
     fame
    -0.84
     apocalyptic
    -0.80
     immortality
    -0.80
     caric
    -0.73
    Die
    -0.72
     shri
    -0.72
     glory
    -0.72
     Communism
    -0.71
     blasp
    -0.71
     doom
    -0.71
    POSITIVE LOGITS
     briefed
    1.16
     consultations
    1.05
     lia
    0.94
     reviewed
    0.90
     contingency
    0.90
     consulted
    0.88
     consultation
    0.86
     discussions
    0.85
     reviewing
    0.85
     inquiries
    0.84
    Activations Density 4.663%

    No Known Activations