© 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. 8-RES_POST_32K-OAI
    11. 6655
    Prev
    Next
    INDEX
    Explanations

    expressions related to political backing and support

    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
    49
    +0.12
    0.5%
    46
    +0.10
    0.5%
    361
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    572
    +0.12
    0.10
    121
    +0.10
    0.09
    608
    +0.09
    0.12
    Negative Logits
    genre
    -0.67
     rand
    -0.59
    artifacts
    -0.58
    elligent
    -0.57
     scaven
    -0.57
     quir
    -0.57
    uti
    -0.57
     exits
    -0.56
    andi
    -0.56
    odd
    -0.55
    POSITIVE LOGITS
     whom
    0.77
     Lifetime
    0.75
    pload
    0.75
    soDeliveryDate
    0.74
    âĹ¼
    0.74
     Galile
    0.71
    SourceFile
    0.71
     Screw
    0.68
     Its
    0.67
     [|
    0.67
    Activations Density 2.472%

    No Known Activations