© 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. 10-RES_POST_32K-OAI
    11. 28961
    Prev
    Next
    INDEX
    Explanations

    references to specific military equipment and their variants

    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
    508
    +0.10
    0.4%
    527
    +0.10
    0.4%
    230
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    373
    +0.10
    -0.02
    422
    +0.10
    0.01
    326
    +0.08
    0.03
    Negative Logits
    krit
    -0.73
    enegger
    -0.70
     "$:/
    -0.68
     Survivors
    -0.68
    eenth
    -0.68
     Lump
    -0.67
    thood
    -0.66
     Olivier
    -0.65
    ecause
    -0.64
    Merit
    -0.64
    POSITIVE LOGITS
    trl
    0.68
    hub
    0.63
     McCarthy
    0.61
    NK
    0.61
    axy
    0.58
    £ı
    0.58
    config
    0.57
     awoken
    0.56
    upper
    0.56
    ESH
    0.55
    Activations Density 0.140%

    No Known Activations