© 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
    3. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Attention Out - 32k
    9. 3-ATT_32K-OAI
    11. 3600
    Prev
    Next
    INDEX
    Explanations

    terms related to legal and official proceedings

    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
    466
    +0.10
    0.4%
    104
    +0.09
    0.4%
    500
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    266
    +0.10
    0.34
    99
    +0.09
    0.10
    101
    +0.09
    0.08
    Negative Logits
     flu
    -0.79
     ponds
    -0.79
     iceberg
    -0.75
     summers
    -0.74
     exploding
    -0.73
     vitamins
    -0.72
    flu
    -0.72
     leaps
    -0.72
    ritch
    -0.72
     seams
    -0.71
    POSITIVE LOGITS
     decree
    0.95
    ername
    0.95
     authorizing
    0.91
     Honour
    0.91
    ij士
    0.89
     appoint
    0.88
    uties
    0.84
     proclamation
    0.84
    assad
    0.83
     Memor
    0.83
    Activations Density 2.317%

    No Known Activations