© 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. 8-RES_POST_32K-OAI
    11. 6397
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "my."

    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
    132
    +0.12
    0.6%
    702
    +0.10
    0.5%
    645
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    385
    +0.12
    0.06
    362
    +0.10
    0.05
    62
    +0.10
    0.04
    Negative Logits
     ACTIONS
    -0.68
    \/\/
    -0.65
    lishes
    -0.65
    utters
    -0.62
    MpServer
    -0.61
    ensis
    -0.61
     Lauder
    -0.60
    ãĤĭ
    -0.60
     Breach
    -0.58
    itud
    -0.58
    POSITIVE LOGITS
     takeaway
    0.94
     reason
    0.92
     guess
    0.90
     apologies
    0.89
     thing
    0.87
     goal
    0.84
     condolences
    0.84
     God
    0.82
    self
    0.82
    stery
    0.79
    Activations Density 0.213%

    No Known Activations