© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    3. GPT2-Small
    4. Attention Out - 32k
    5. 3-ATT_32K-OAI
    6. 4172
    Prev
    Next
    INDEX
    Explanations

    terms related to astronomy and atmospheric phenomena, particularly the northern lights and associated visual experiences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-att_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-attn-out-SAEs/v5_32k_layer_3
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.3.hook_attn_out
    Hook Layer
    3
    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
    538
    +0.13
    0.6%
    613
    +0.12
    0.5%
    713
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    447
    +0.13
    0.27
    713
    +0.12
    0.11
    107
    +0.11
    0.10
    Negative Logits
    ynes
    -0.81
    ority
    -0.77
    TeX
    -0.76
    ilies
    -0.74
    ci
    -0.73
    artisan
    -0.73
    cule
    -0.73
    hers
    -0.71
    tery
    -0.71
    ilit
    -0.71
    POSITIVE LOGITS
     Display
    0.74
     Clear
    0.74
     retri
    0.71
     display
    0.69
    Images
    0.68
     announ
    0.67
     VIS
    0.64
     artif
    0.64
     emitting
    0.64
     Lumpur
    0.63
    Activations Density 15.319%

    No Known Activations