© 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. 3827
    Prev
    Next
    INDEX
    Explanations

    concepts related to accountability and responsibility for actions

    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
    586
    +0.12
    0.5%
    457
    +0.11
    0.5%
    653
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    586
    +0.12
    0.04
    653
    +0.11
    0.04
    531
    +0.10
    0.04
    Negative Logits
    vine
    -0.77
    hern
    -0.76
    quire
    -0.69
    aquin
    -0.68
    bern
    -0.66
    bish
    -0.66
    emis
    -0.66
    aters
    -0.66
    jen
    -0.65
    replace
    -0.65
    POSITIVE LOGITS
     challeng
    0.76
     concess
    0.75
     precarious
    0.73
     tight
    0.71
     conduc
    0.70
     hostage
    0.70
     indefinitely
    0.69
     ledge
    0.68
     holds
    0.68
     stake
    0.68
    Activations Density 0.556%

    No Known Activations