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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 13544
    Prev
    Next
    INDEX
    Explanations

    references to magma in geological contexts

    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
    270
    +0.15
    1.0%
    649
    +0.14
    1.0%
    160
    +0.14
    1.0%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    93
    +0.15
    0.02
    701
    +0.14
    0.01
    324
    +0.14
    0.01
    Negative Logits
    ãĤª
    -0.86
    ãĤ¤ãĥĪ
    -0.83
    ãĥ©
    -0.79
     Communications
    -0.73
     Sek
    -0.72
     Kak
    -0.70
     edit
    -0.70
    PsyNetMessage
    -0.70
    ãĤ¶
    -0.70
     Luxem
    -0.70
    POSITIVE LOGITS
    estones
    0.89
    artifacts
    0.88
    ologies
    0.88
    otypes
    0.87
    etheus
    0.86
    abilities
    0.86
    otype
    0.84
    ucl
    0.84
    vable
    0.83
    uctor
    0.83
    Activations Density 0.039%

    No Known Activations