© 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. 5318
    Prev
    Next
    INDEX
    Explanations

    references to instinct and intuition

    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
    635
    +0.13
    0.6%
    140
    +0.13
    0.6%
    285
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    285
    +0.13
    0.03
    317
    +0.13
    0.04
    199
    +0.11
    0.04
    Negative Logits
    jri
    -0.71
    ouk
    -0.71
    iatus
    -0.70
    paces
    -0.70
    eday
    -0.69
    qa
    -0.69
    acters
    -0.65
    jan
    -0.64
    gre
    -0.64
    itary
    -0.64
    POSITIVE LOGITS
     instinct
    1.26
     instinctively
    1.08
     instincts
    1.07
     reaction
    0.97
     reflex
    0.95
     impulses
    0.90
    ual
    0.89
     impulse
    0.85
    arily
    0.84
     intuition
    0.83
    Activations Density 0.373%

    No Known Activations