© 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. 29237
    Prev
    Next
    INDEX
    Explanations

    references to significant events or issues related to public health and safety

    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
    71
    +0.11
    0.5%
    246
    +0.10
    0.5%
    85
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    85
    +0.11
    0.10
    408
    +0.10
    0.16
    1
    +0.10
    0.08
    Negative Logits
    uana
    -0.83
    vantage
    -0.79
    auna
    -0.73
    arity
    -0.70
    erion
    -0.68
    epad
    -0.68
    undrum
    -0.67
    iciency
    -0.65
    ouf
    -0.63
    ilty
    -0.63
    POSITIVE LOGITS
     outings
    1.52
     incidents
    1.49
     occasions
    1.45
     trips
    1.44
     victories
    1.43
     defeats
    1.43
     encounters
    1.42
     disasters
    1.41
     setbacks
    1.41
     bouts
    1.40
    Activations Density 3.957%

    No Known Activations