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

    Neuronpedia

    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. 4521
    Prev
    Next
    INDEX
    Explanations

    URLs and web-related links

    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
    613
    +0.13
    0.6%
    85
    +0.12
    0.5%
    250
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    284
    +0.13
    0.03
    640
    +0.12
    0.04
    314
    +0.10
    -0.01
    Negative Logits
     Michelle
    -0.80
     Lauder
    -0.73
     Melania
    -0.71
     Bezos
    -0.68
     enh
    -0.68
     Jill
    -0.68
     Siem
    -0.67
     Dorothy
    -0.67
     Nieto
    -0.66
    ograp
    -0.66
    POSITIVE LOGITS
    thread
    1.19
    topic
    1.08
    forums
    1.07
    forum
    1.05
    ussion
    1.02
    Thread
    0.97
     forum
    0.97
     Forums
    0.96
     thread
    0.95
     threads
    0.92
    Activations Density 0.327%

    No Known Activations