© 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. 3654
    Prev
    Next
    INDEX
    Explanations

    different forms and contexts of the word "expression."

    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
    215
    +0.11
    0.5%
    487
    +0.11
    0.5%
    463
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    679
    +0.11
    0.06
    759
    +0.11
    0.04
    507
    +0.11
    0.03
    Negative Logits
     raid
    -0.73
    isites
    -0.68
     raiding
    -0.68
     CDs
    -0.64
     exploits
    -0.64
     Bagg
    -0.63
    usercontent
    -0.62
     ammunition
    -0.61
    yip
    -0.61
     CTR
    -0.61
    POSITIVE LOGITS
    Ĭ±
    0.73
     José
    0.67
    jab
    0.66
    Tags
    0.65
    Reading
    0.65
     horizont
    0.64
    Graph
    0.64
    ëĭ
    0.63
    Cola
    0.62
    ability
    0.62
    Activations Density 0.822%

    No Known Activations