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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    3. GPT2-Small
    4. Residual Post - 32k
    5. 8-RES_POST_32K-OAI
    6. 5197
    Prev
    Next
    INDEX
    Explanations

    terms related to class definitions in programming languages

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_8.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.8.hook_resid_post
    Hook Layer
    8
    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
    679
    +0.12
    0.5%
    167
    +0.11
    0.5%
    616
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    167
    +0.12
    0.01
    679
    +0.11
    0.02
    42
    +0.11
    0.02
    Negative Logits
    igators
    -0.70
    andum
    -0.68
    OPLE
    -0.67
    hower
    -0.65
     gloom
    -0.65
     Mile
    -0.65
     gull
    -0.63
     outdoors
    -0.63
    vous
    -0.62
    nesota
    -0.62
    POSITIVE LOGITS
    names
    1.07
    name
    0.94
    Loader
    0.93
    path
    0.92
    ortium
    0.91
    ifier
    0.87
    loader
    0.83
    feature
    0.82
    Name
    0.81
    ification
    0.78
    Activations Density 0.085%

    No Known Activations