© 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. 28233
    Prev
    Next
    INDEX
    Explanations

    instructions or requests for taking action related to reading and checking information

    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
    273
    +0.12
    0.5%
    467
    +0.10
    0.5%
    720
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    273
    +0.12
    0.05
    611
    +0.10
    0.05
    172
    +0.10
    0.04
    Negative Logits
    ãĤ¼ãĤ¦ãĤ¹
    -0.99
    aga
    -0.69
    inois
    -0.62
    inatory
    -0.62
    folk
    -0.61
    awks
    -0.61
    ilon
    -0.58
     paralyzed
    -0.58
    uries
    -0.56
    orescence
    -0.56
    POSITIVE LOGITS
     FAQ
    0.96
     http
    0.91
     below
    0.90
     https
    0.89
    endum
    0.85
     instructions
    0.83
     carefully
    0.82
     youtube
    0.82
     section
    0.81
     READ
    0.80
    Activations Density 0.461%

    No Known Activations