© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 5547
    Prev
    Next
    INDEX
    Explanations

    references to arms in various contexts

    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
    595
    +0.10
    0.4%
    356
    +0.10
    0.4%
    394
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    516
    +0.10
    0.02
    463
    +0.10
    0.02
    553
    +0.10
    0.02
    Negative Logits
    actionDate
    -0.77
     Sound
    -0.75
     Tycoon
    -0.71
     Vide
    -0.69
     Outbreak
    -0.64
     Atmosp
    -0.64
     Commission
    -0.61
     DevOnline
    -0.61
    lucent
    -0.60
     Ud
    -0.60
    POSITIVE LOGITS
    chair
    1.30
    aments
    1.24
    ament
    1.11
    ageddon
    1.10
    chairs
    1.06
    strong
    0.99
    oire
    0.93
     wrestle
    0.93
    lift
    0.87
    guards
    0.84
    Activations Density 0.056%

    No Known Activations