© 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. 3414
    Prev
    Next
    INDEX
    Explanations

    statistical data and references to studies or reports

    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
    554
    +0.12
    0.6%
    613
    +0.10
    0.5%
    479
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    148
    +0.12
    0.13
    193
    +0.10
    0.10
    480
    +0.10
    0.12
    Negative Logits
     floats
    -0.69
     races
    -0.68
     choke
    -0.67
     rapes
    -0.67
     gull
    -0.66
     sleeps
    -0.63
    asts
    -0.62
    isol
    -0.61
     stalls
    -0.61
     assassinated
    -0.60
    POSITIVE LOGITS
    WARE
    0.80
    FTWARE
    0.77
    ERC
    0.76
    è»
    0.72
     Payment
    0.72
    IRC
    0.72
    âĹ¼
    0.71
    accompanied
    0.71
     Replacement
    0.70
    ï¸
    0.67
    Activations Density 11.380%

    No Known Activations