© 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. 32302
    Prev
    Next
    INDEX
    Explanations

    references to channels, particularly in the context of media or communication

    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
    288
    +0.14
    0.6%
    455
    +0.13
    0.6%
    271
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.14
    0.03
    41
    +0.13
    0.02
    459
    +0.11
    0.01
    Negative Logits
    enhagen
    -0.76
    berman
    -0.75
    hiro
    -0.70
     Gloria
    -0.66
    hya
    -0.66
     Arkham
    -0.65
    stad
    -0.62
    gling
    -0.62
     Raider
    -0.62
     Hots
    -0.61
    POSITIVE LOGITS
     channel
    0.82
    lain
    0.82
     channels
    0.82
    annels
    0.79
     blockers
    0.78
    Fire
    0.77
    agall
    0.75
    washer
    0.74
     surfing
    0.74
    izen
    0.72
    Activations Density 0.142%

    No Known Activations