© 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. 30377
    Prev
    Next
    INDEX
    Explanations

    instances of the word "used" and related phrases indicating applications or implementations of actions or strategies

    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
    762
    +0.10
    0.4%
    756
    +0.10
    0.4%
    141
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    225
    +0.10
    0.09
    762
    +0.10
    0.10
    386
    +0.09
    0.08
    Negative Logits
    bilt
    -0.73
     guiName
    -0.67
     footprints
    -0.63
     originals
    -0.61
     Plymouth
    -0.60
     Indigo
    -0.60
    accompan
    -0.59
    Located
    -0.59
     CFR
    -0.59
     flats
    -0.59
    POSITIVE LOGITS
     against
    1.55
    against
    1.49
     Against
    1.15
     leverage
    0.98
    Against
    0.98
     AGA
    0.94
     rhet
    0.91
     blackmail
    0.91
    weapon
    0.88
     defensively
    0.84
    Activations Density 1.211%

    No Known Activations