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. 13538
    Prev
    Next
    INDEX
    Explanations

    numerical values and statistics

    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
    578
    +0.30
    2.3%
    339
    +0.22
    1.7%
    85
    +0.17
    1.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.30
    0.01
    151
    +0.22
    0.01
    587
    +0.17
    0.01
    Negative Logits
    worm
    -0.77
    AMY
    -0.76
    URE
    -0.75
    abad
    -0.74
    AIN
    -0.73
    Rod
    -0.73
    OWS
    -0.70
    PATH
    -0.70
    ãĤ´
    -0.70
    ANGE
    -0.70
    POSITIVE LOGITS
     newcom
    0.99
    ortunately
    0.91
     hypot
    0.86
    paio
    0.84
    bil
    0.82
     predec
    0.81
     tremend
    0.80
     cannabin
    0.79
     clipboard
    0.79
    uckland
    0.79
    Activations Density 0.011%

    No Known Activations