© 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. 10-RES_POST_32K-OAI
    11. 30611
    Prev
    Next
    INDEX
    Explanations

    phrases indicating direct quotations or reported speech in text

    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
    428
    +0.11
    0.5%
    373
    +0.11
    0.5%
    410
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    224
    +0.11
    0.09
    346
    +0.11
    0.08
    291
    +0.10
    0.07
    Negative Logits
     depends
    -0.74
    bred
    -0.70
     Became
    -0.70
    isable
    -0.69
     belongs
    -0.68
     Designed
    -0.64
    nir
    -0.64
     migrated
    -0.63
     depended
    -0.62
    tain
    -0.61
    POSITIVE LOGITS
     saying
    1.71
     telling
    1.38
     stating
    1.37
     Saying
    1.29
     remark
    1.24
     arguing
    1.22
     claiming
    1.18
     insisting
    1.17
     praising
    1.17
     noting
    1.16
    Activations Density 1.304%

    No Known Activations