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. 13588
    Prev
    Next
    INDEX
    Explanations

    terms related to software, technical specifications, and computing environments

    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
    587
    +0.17
    1.1%
    376
    +0.16
    1.0%
    644
    +0.15
    1.0%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    107
    +0.17
    0.01
    27
    +0.16
    0.01
    15
    +0.15
    0.01
    Negative Logits
    代
    -0.91
    ãĥ³ãĤ¸
    -0.84
    theless
    -0.78
    ãĥį
    -0.77
    ãĤ¯
    -0.76
    ãĤ·ãĥ£
    -0.76
    é¾įå
    -0.73
    ãĥ¼ãĥĨ
    -0.72
    ãĤ®
    -0.72
    UGC
    -0.70
    POSITIVE LOGITS
    rehend
    1.17
    brids
    0.90
    retty
    0.86
    estern
    0.85
    galitarian
    0.84
    orld
    0.83
    hement
    0.82
    overe
    0.78
    Introduced
    0.78
    vae
    0.78
    Activations Density 0.027%

    No Known Activations