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. 13567
    Prev
    Next
    INDEX
    Explanations

    references to specific places, particularly street names or locations

    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
    213
    +0.30
    2.3%
    59
    +0.22
    1.7%
    146
    +0.18
    1.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    342
    +0.30
    0.00
    365
    +0.22
    0.01
    629
    +0.18
    0.00
    Negative Logits
    interstitial
    -0.85
    ungle
    -0.81
    wcs
    -0.81
    inventoryQuantity
    -0.77
    xtap
    -0.75
    */(
    -0.75
    othal
    -0.74
    zees
    -0.73
    acular
    -0.73
    ongyang
    -0.72
    POSITIVE LOGITS
    mond
    0.84
    cloth
    0.84
    tons
    0.83
    ton
    0.82
    essa
    0.81
    driver
    0.81
    vale
    0.81
    Nap
    0.80
    nt
    0.80
    frey
    0.79
    Activations Density 0.008%

    No Known Activations