© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 5492
    Prev
    Next
    INDEX
    Explanations

    mentions of specific individuals

    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
    525
    +0.13
    0.6%
    84
    +0.10
    0.5%
    723
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.13
    0.02
    525
    +0.10
    0.01
    742
    +0.10
    0.01
    Negative Logits
    ele
    -0.87
    inals
    -0.80
    ally
    -0.79
    tons
    -0.78
    inally
    -0.77
    unity
    -0.73
    ings
    -0.71
    ered
    -0.70
    erences
    -0.69
    alty
    -0.67
    POSITIVE LOGITS
    uria
    0.73
     Parenthood
    0.71
     orbits
    0.67
    borg
    0.66
    gem
    0.65
    stadt
    0.64
    jin
    0.64
    GO
    0.63
    gang
    0.63
     mathemat
    0.63
    Activations Density 0.055%

    No Known Activations