© 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. 8-RES_POST_32K-OAI
    11. 7878
    Prev
    Next
    INDEX
    Explanations

    phrases that indicate the presence of a significant event or situation

    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
    582
    +0.12
    0.6%
    241
    +0.12
    0.6%
    62
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    159
    +0.12
    0.11
    598
    +0.12
    0.06
    186
    +0.11
    0.06
    Negative Logits
     facult
    -0.72
     corpus
    -0.67
     deterrent
    -0.64
     princip
    -0.61
     ful
    -0.61
     harmful
    -0.59
     restrictive
    -0.59
    ailable
    -0.59
     repressive
    -0.58
     Comprehensive
    -0.57
    POSITIVE LOGITS
     isn
    0.98
     wasn
    0.93
     ain
    0.86
     sucks
    0.77
     definitely
    0.76
     is
    0.73
     hasn
    0.73
     reminds
    0.73
     doesn
    0.73
    nesday
    0.70
    Activations Density 0.230%

    No Known Activations