Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    3. GPT2-Small
    4. Attention Out - 128k
    5. 0-ATT_128K-OAI
    6. 6467
    Prev
    Next
    INDEX
    Explanations

    phrases related to burdens and the act of imposing

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    jbloom/GPT2-Small-OAI-v5-128k-attn-out-SAEs/v5_128k_layer_0
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    131,072
    Data Type
    torch.float32
    Hook Name
    blocks.0.hook_attn_out
    Hook Layer
    0
    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
    167
    +0.15
    1.2%
    213
    +0.15
    1.2%
    386
    +0.15
    1.1%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    151
    +0.15
    0.02
    85
    +0.15
    0.02
    653
    +0.15
    0.02
    Negative Logits
     Franch
    -0.84
     Brav
    -0.82
    athan
    -0.81
     Ples
    -0.80
    20439
    -0.77
    2019
    -0.76
     Tie
    -0.74
     Stev
    -0.73
     WARRANT
    -0.73
     Lub
    -0.72
    POSITIVE LOGITS
    ¥ŀ
    1.19
    ascript
    1.03
    alore
    1.00
    athered
    0.99
    anooga
    0.98
     cumbers
    0.97
     glim
    0.96
     citiz
    0.95
     nomine
    0.94
    regor
    0.93
    Activations Density 0.043%

    No Known Activations