© 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. 10-RES_POST_32K-OAI
    11. 31930
    Prev
    Next
    INDEX
    Explanations

    repeated instances of the word "been" in various contexts

    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
    327
    +0.11
    0.5%
    557
    +0.09
    0.4%
    318
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    393
    +0.11
    0.09
    608
    +0.09
    0.08
    573
    +0.09
    0.03
    Negative Logits
    éĹĺ
    -0.73
     tomorrow
    -0.70
     Unsure
    -0.67
     anymore
    -0.66
    }}}
    -0.65
     rematch
    -0.64
    ļéĨĴ
    -0.62
    ordan
    -0.61
    later
    -0.60
    terday
    -0.60
    POSITIVE LOGITS
     steadily
    1.07
     steady
    1.05
     lately
    1.02
     since
    0.98
     consistently
    0.93
     consistent
    0.93
     dogged
    0.86
     success
    0.80
     steadfast
    0.79
    since
    0.78
    Activations Density 1.214%

    No Known Activations