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. 6462
    Prev
    Next
    INDEX
    Explanations

    numerical data and punctuation

    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
    116
    +0.18
    1.1%
    57
    +0.16
    1.0%
    417
    +0.15
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    556
    +0.18
    0.04
    257
    +0.16
    0.04
    640
    +0.15
    0.04
    Negative Logits
     compr
    -0.81
     exha
    -0.78
     metic
    -0.77
     tremend
    -0.76
     tofu
    -0.75
    ername
    -0.73
    wagen
    -0.71
    quist
    -0.71
     vague
    -0.71
     occas
    -0.70
    POSITIVE LOGITS
    iHUD
    0.93
    CVE
    0.93
     Mehran
    0.91
    "]=>
    0.85
    UGC
    0.83
    ],[
    0.82
    dylib
    0.81
    iPhone
    0.81
    wcsstore
    0.78
    MpServer
    0.78
    Activations Density 0.055%

    No Known Activations