© 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
    3. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Attention Out - 32k
    9. 3-ATT_32K-OAI
    11. 4387
    Prev
    Next
    INDEX
    Explanations

    concepts related to disabling features or functionalities in technology

    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
    509
    +0.11
    0.5%
    694
    +0.09
    0.4%
    192
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    694
    +0.11
    0.02
    255
    +0.09
    0.03
    192
    +0.09
    0.02
    Negative Logits
    lished
    -0.83
    代
    -0.81
     Proced
    -0.74
    esta
    -0.72
    ixt
    -0.71
     Zup
    -0.70
    PsyNetMessage
    -0.69
    gt
    -0.69
    undreds
    -0.69
     somewhere
    -0.68
    POSITIVE LOGITS
    loading
    0.84
    activation
    0.79
     protections
    0.72
     eligibility
    0.71
     blackout
    0.71
     lockout
    0.70
     mobility
    0.70
     automatic
    0.68
     blindness
    0.67
     exclusion
    0.67
    Activations Density 0.241%

    No Known Activations