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

    Neuronpedia

    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. 29619
    Prev
    Next
    INDEX
    Explanations

    sentences that express definitive statements or conclusions

    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
    720
    +0.13
    0.6%
    287
    +0.10
    0.5%
    431
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    6
    +0.13
    0.32
    145
    +0.10
    0.24
    622
    +0.09
    0.19
    Negative Logits
     netted
    -0.81
     vel
    -0.76
    united
    -0.73
     shipped
    -0.72
     landfall
    -0.72
     000000
    -0.70
     ported
    -0.69
     ozone
    -0.69
    wald
    -0.68
     irrad
    -0.68
    POSITIVE LOGITS
     Volunte
    1.13
     Anger
    1.09
     Adults
    1.04
     Anxiety
    1.02
     bullies
    1.01
     Psychological
    1.01
     Behavior
    1.01
    verbal
    0.97
     counselors
    0.96
     behaviors
    0.95
    Activations Density 2.052%

    No Known Activations