© 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. 26919
    Prev
    Next
    INDEX
    Explanations

    actions related to financial transactions or economic activities

    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
    218
    +0.12
    0.5%
    586
    +0.10
    0.5%
    483
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    750
    +0.12
    0.07
    366
    +0.10
    0.05
    468
    +0.09
    0.06
    Negative Logits
    rian
    -0.63
    abad
    -0.59
    nard
    -0.59
    ominated
    -0.57
    bern
    -0.57
    chers
    -0.56
    rophic
    -0.56
    athom
    -0.55
    igma
    -0.55
    è¦ļéĨĴ
    -0.54
    POSITIVE LOGITS
     off
    2.32
    off
    1.83
     Off
    1.74
    Off
    1.63
    offs
    1.63
     OFF
    1.51
    OFF
    1.50
     offline
    0.78
     apart
    0.71
     away
    0.70
    Activations Density 1.801%

    No Known Activations