© 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. 7997
    Prev
    Next
    INDEX
    Explanations

    mentions of specific individuals or entities, particularly in a formal or notable context

    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
    394
    +0.13
    0.6%
    492
    +0.12
    0.6%
    373
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.13
    0.03
    343
    +0.12
    0.02
    57
    +0.12
    0.01
    Negative Logits
    itarian
    -0.68
     Origin
    -0.68
    ãĥ¯ãĥ³
    -0.67
     Nobel
    -0.64
     Radius
    -0.64
     Scand
    -0.64
    Production
    -0.61
    rified
    -0.61
     distribution
    -0.60
     Bauer
    -0.59
    POSITIVE LOGITS
    atoon
    0.96
    eters
    0.88
    ggles
    0.87
    utes
    0.86
    uten
    0.86
    uble
    0.82
    ggle
    0.81
    ute
    0.81
    ignt
    0.79
    zers
    0.78
    Activations Density 0.100%

    No Known Activations