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. 6475
    Prev
    Next
    INDEX
    Explanations

    references to the state of Uttar Pradesh

    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
    23
    +0.18
    1.4%
    707
    +0.16
    1.2%
    739
    +0.15
    1.1%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    538
    +0.18
    0.01
    288
    +0.16
    0.02
    447
    +0.15
    0.01
    Negative Logits
    pred
    -0.74
    chart
    -0.71
    aneers
    -0.71
    rend
    -0.70
    focus
    -0.68
    HER
    -0.68
    aer
    -0.67
    offensive
    -0.67
    Assembly
    -0.66
    wings
    -0.65
    POSITIVE LOGITS
     Lumpur
    1.05
    utical
    0.85
    ivalent
    0.79
    Ĥª
    0.79
    Ħ¢
    0.78
    earances
    0.77
    umbai
    0.75
     "$:/
    0.74
    awaru
    0.72
    Ĥİ
    0.71
    Activations Density 0.020%

    No Known Activations