Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    3. GPT2-Small
    4. Attention Out - 128k
    5. 0-ATT_128K-OAI
    6. 13546
    Prev
    Next
    INDEX
    Explanations

    occurrences of the name "Speier"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    jbloom/GPT2-Small-OAI-v5-128k-attn-out-SAEs/v5_128k_layer_0
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    131,072
    Data Type
    torch.float32
    Hook Name
    blocks.0.hook_attn_out
    Hook Layer
    0
    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
    216
    +0.25
    2.0%
    679
    +0.22
    1.7%
    195
    +0.21
    1.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    538
    +0.25
    0.01
    505
    +0.22
    0.01
    439
    +0.21
    0.01
    Negative Logits
    è£ħ
    -0.88
    apy
    -0.79
     Pradesh
    -0.73
     è£ıç
    -0.73
    ãĤĭ
    -0.72
    ãģ®
    -0.71
    uras
    -0.71
    ption
    -0.71
    etheless
    -0.71
    ãĥŁ
    -0.71
    POSITIVE LOGITS
    emark
    0.85
    cius
    0.82
    inventoryQuantity
    0.78
    iquid
    0.78
    etter
    0.77
    eport
    0.75
    ebus
    0.74
    eway
    0.73
    natureconservancy
    0.70
    eton
    0.70
    Activations Density 0.005%

    No Known Activations