© 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. 4519
    Prev
    Next
    INDEX
    Explanations

    environmental issues and their consequences

    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
    136
    +0.10
    0.4%
    694
    +0.09
    0.4%
    459
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.10
    0.15
    225
    +0.09
    0.12
    258
    +0.09
    0.12
    Negative Logits
    20439
    -0.78
    witz
    -0.62
    ãĥĦ
    -0.59
    WithNo
    -0.59
    izont
    -0.58
    racuse
    -0.57
    ãĥ¼ãĥĨãĤ£
    -0.57
    kb
    -0.56
    ãĤ§
    -0.56
    lim
    -0.56
    POSITIVE LOGITS
     etc
    0.91
    etc
    0.86
    anything
    0.76
    âĢ¦.
    0.76
     all
    0.74
    âĢ¦
    0.72
     âĢ¦
    0.71
    âĢ¦)
    0.71
    all
    0.69
     ALL
    0.68
    Activations Density 2.748%

    No Known Activations