© 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. 27264
    Prev
    Next
    INDEX
    Explanations

    instances of the word "was" and phrases indicating state or identity

    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
    55
    +0.12
    0.5%
    218
    +0.12
    0.5%
    102
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    102
    +0.12
    0.17
    55
    +0.12
    0.15
    566
    +0.11
    0.11
    Negative Logits
    esm
    -0.77
    ital
    -0.72
    izoph
    -0.71
     beginnings
    -0.70
     bindings
    -0.70
     spheres
    -0.67
    tnc
    -0.66
     Supports
    -0.65
    issues
    -0.65
    urgy
    -0.64
    POSITIVE LOGITS
     unaware
    0.95
     deceased
    0.94
     identified
    0.84
     himself
    0.82
     unwilling
    0.80
     someone
    0.80
     appointed
    0.80
     blinded
    0.79
     asleep
    0.79
     rewarded
    0.79
    Activations Density 1.105%

    No Known Activations