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. 13558
    Prev
    Next
    INDEX
    Explanations

    expressions of uncertainty or skepticism

    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
    247
    +0.21
    1.4%
    718
    +0.14
    1.0%
    624
    +0.14
    1.0%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    356
    +0.21
    0.01
    52
    +0.14
    0.01
    99
    +0.14
    0.02
    Negative Logits
    ocations
    -0.91
    estones
    -0.83
    ocation
    -0.72
     Contents
    -0.71
     Contains
    -0.70
    estone
    -0.70
    IU
    -0.70
    NetMessage
    -0.69
    ickr
    -0.69
    itionally
    -0.69
    POSITIVE LOGITS
    lessly
    1.02
    zzle
    0.77
    iffe
    0.74
    lessness
    0.71
    stellar
    0.71
    taker
    0.71
     doubt
    0.70
    ministic
    0.68
    lust
    0.67
     internally
    0.66
    Activations Density 0.296%

    No Known Activations