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. 6458
    Prev
    Next
    INDEX
    Explanations

    mentions of sexual assault and related issues

    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
    643
    +0.17
    1.2%
    679
    +0.16
    1.2%
    438
    +0.15
    1.1%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    417
    +0.17
    0.01
    116
    +0.16
    0.01
    130
    +0.15
    0.01
    Negative Logits
     Pwr
    -0.79
    soDeliveryDate
    -0.78
     sidx
    -0.77
    PF
    -0.76
    iasm
    -0.74
    STON
    -0.73
    âĶĢâĶĢ
    -0.73
    heric
    -0.70
    ãĥ¼ãĥ«
    -0.70
    YP
    -0.70
    POSITIVE LOGITS
    ariat
    0.77
     whistlebl
    0.74
     assassinated
    0.72
    anamo
    0.68
     Appeals
    0.67
     Oversight
    0.67
     subpoen
    0.65
     detectives
    0.65
    raq
    0.65
     investigator
    0.65
    Activations Density 0.009%

    No Known Activations