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. 6435
    Prev
    Next
    INDEX
    Explanations

    expressions of gratitude

    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
    679
    +0.31
    2.2%
    566
    +0.21
    1.5%
    151
    +0.18
    1.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    415
    +0.31
    0.02
    151
    +0.21
    0.01
    302
    +0.18
    0.02
    Negative Logits
    女
    -1.07
     Adin
    -0.82
     Mamm
    -0.80
     Scotia
    -0.78
     Wolves
    -0.75
     Birch
    -0.75
     Borough
    -0.73
    idia
    -0.67
     Fey
    -0.66
    idad
    -0.65
    POSITIVE LOGITS
    chwitz
    1.05
    awaru
    1.00
    ¿½
    0.93
    uliffe
    0.84
    uristic
    0.83
    ging
    0.82
    ivating
    0.79
    itialized
    0.78
    isec
    0.77
    graded
    0.77
    Activations Density 0.012%

    No Known Activations