© 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. 6876
    Prev
    Next
    INDEX
    Explanations

    occurrences of the pronouns "I," "we," and "it," indicating a focus on personal or collective experiences

    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
    314
    +0.15
    0.7%
    418
    +0.12
    0.5%
    759
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    354
    +0.15
    0.08
    697
    +0.12
    0.06
    287
    +0.09
    0.06
    Negative Logits
     */
    -0.85
     listed
    -0.68
    .''.
    -0.64
     ÏĦ
    -0.64
     �
    -0.64
     RS
    -0.63
     âĢº
    -0.63
    .","
    -0.63
    -0.60
     âĶĤ
    -0.59
    POSITIVE LOGITS
    odore
    1.21
    usterity
    1.13
    gency
    1.10
    resa
    1.08
    tenance
    1.08
    xiety
    1.08
    alyst
    1.00
    ntil
    0.98
    withstanding
    0.98
    mosp
    0.91
    Activations Density 0.606%

    No Known Activations