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. 6454
    Prev
    Next
    INDEX
    Explanations

    words related to inquisitiveness or curiosity

    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
    622
    +0.18
    1.2%
    383
    +0.16
    1.1%
    201
    +0.14
    1.0%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    390
    +0.18
    0.01
    763
    +0.16
    0.01
    0
    +0.14
    0.01
    Negative Logits
    odi
    -0.92
    engeance
    -0.91
    akuya
    -0.88
    iciency
    -0.86
    ixel
    -0.85
     tremend
    -0.84
    velength
    -0.84
    ortium
    -0.83
    rha
    -0.82
    annie
    -0.82
    POSITIVE LOGITS
    ĸļ
    0.87
    forth
    0.82
    ãĤ¹ãĥĪ
    0.77
    EStreamFrame
    0.73
    ļéĨĴ
    0.70
    gerald
    0.69
    foot
    0.68
     territorial
    0.67
    ioned
    0.67
    ¼
    0.66
    Activations Density 0.003%

    No Known Activations