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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsBlogSlackPrivacy & 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. 13568
    Prev
    Next
    INDEX
    Explanations

    technical terms related to programming and web development

    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
    167
    +0.15
    1.1%
    503
    +0.15
    1.1%
    644
    +0.15
    1.1%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.15
    0.00
    149
    +0.15
    0.00
    447
    +0.15
    0.00
    Negative Logits
     Trials
    -0.87
     Span
    -0.87
     Goods
    -0.85
     Mercury
    -0.85
     Trave
    -0.85
     Prayer
    -0.80
     Grain
    -0.79
     Cheong
    -0.79
     Guinea
    -0.76
    moon
    -0.75
    POSITIVE LOGITS
    eatures
    1.25
    tymology
    1.19
     unden
    1.07
     tremend
    1.07
     destro
    1.07
    jriwal
    1.03
     volunte
    1.01
     skelet
    1.01
     challeng
    1.01
     enthusi
    1.00
    Activations Density 0.005%

    No Known Activations