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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 3828
    Prev
    Next
    INDEX
    Explanations

    references to social media platforms and interaction prompts

    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
    525
    +0.11
    0.5%
    750
    +0.10
    0.5%
    674
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    525
    +0.11
    0.02
    739
    +0.10
    0.01
    500
    +0.10
    0.02
    Negative Logits
    panel
    -0.82
    OHN
    -0.77
    undown
    -0.77
    urion
    -0.73
    ework
    -0.72
    SEE
    -0.72
    URN
    -0.72
    alky
    -0.71
    par
    -0.69
    etime
    -0.68
    POSITIVE LOGITS
     indo
    0.75
     bubbles
    0.66
     toile
    0.66
     Emir
    0.66
     Champ
    0.65
     brill
    0.65
     crib
    0.64
     prest
    0.63
     Bib
    0.63
     effortlessly
    0.63
    Activations Density 0.092%

    No Known Activations