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. 13562
    Prev
    Next
    INDEX
    Explanations

    error messages and subscription-related terms

    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
    59
    +0.21
    1.3%
    128
    +0.18
    1.2%
    311
    +0.18
    1.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    34
    +0.21
    0.01
    604
    +0.18
    0.02
    134
    +0.18
    0.02
    Negative Logits
    hyde
    -0.89
    acid
    -0.87
    entimes
    -0.79
    Sym
    -0.74
    FIX
    -0.74
    paste
    -0.72
    pad
    -0.71
    acha
    -0.71
    icter
    -0.71
    CLE
    -0.70
    POSITIVE LOGITS
     subscribing
    0.96
    rolet
    0.87
     sqor
    0.82
    eals
    0.80
    kees
    0.79
    lisher
    0.75
    yrs
    0.74
    auri
    0.71
     Loans
    0.69
     Annotations
    0.69
    Activations Density 0.134%

    No Known Activations