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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsBlogSlackPrivacy & 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. 13565
    Prev
    Next
    INDEX
    Explanations

    references to individuals named Linda

    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
    526
    +0.32
    2.6%
    336
    +0.18
    1.4%
    211
    +0.17
    1.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.32
    0.02
    151
    +0.18
    0.02
    447
    +0.17
    0.02
    Negative Logits
    WER
    -0.75
     Prospect
    -0.72
     Citation
    -0.65
     Behavior
    -0.64
    essions
    -0.64
     proble
    -0.61
    avanaugh
    -0.60
     Components
    -0.60
     Catalyst
    -0.60
    peak
    -0.60
    POSITIVE LOGITS
    igl
    0.92
    omore
    0.84
    ific
    0.83
    jah
    0.82
    ochet
    0.82
    emaker
    0.78
    oples
    0.78
    itude
    0.77
    ventory
    0.77
    assing
    0.76
    Activations Density 0.038%

    No Known Activations