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. 6451
    Prev
    Next
    INDEX
    Explanations

    phrases indicating acknowledgment or acceptance of a situation

    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
    270
    +0.22
    1.4%
    624
    +0.18
    1.2%
    288
    +0.15
    1.0%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    589
    +0.22
    0.02
    393
    +0.18
    0.02
    203
    +0.15
    0.01
    Negative Logits
    cca
    -0.84
    awaru
    -0.83
    ulously
    -0.80
    verty
    -0.80
    iago
    -0.77
    aeda
    -0.76
    PLA
    -0.76
    isoft
    -0.76
    asley
    -0.75
    senal
    -0.75
    POSITIVE LOGITS
     Scrolls
    0.77
    rities
    0.75
    skirts
    0.74
     Horses
    0.73
     ANN
    0.71
     Memories
    0.71
     NEO
    0.71
     LET
    0.71
     Unle
    0.70
     Organizations
    0.70
    Activations Density 0.063%

    No Known Activations