© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    3. GPT2-Small
    4. Residual Post - 32k
    5. 10-RES_POST_32K-OAI
    6. 28986
    Prev
    Next
    INDEX
    Explanations

    references to specific weeks or games in a schedule or season

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_10.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    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
    622
    +0.12
    0.5%
    512
    +0.11
    0.5%
    373
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    651
    +0.12
    0.02
    610
    +0.11
    0.01
    26
    +0.10
    0.01
    Negative Logits
     pandemonium
    -0.68
    QUI
    -0.65
     pse
    -0.65
     crystal
    -0.61
     seismic
    -0.60
    PUT
    -0.58
     Morse
    -0.58
    ptin
    -0.58
     sophistic
    -0.57
    æŃ¦
    -0.57
    POSITIVE LOGITS
    eight
    0.94
    ounces
    0.87
    ender
    0.87
    oulder
    0.87
    nine
    0.86
    seven
    0.85
    five
    0.84
    six
    0.83
    three
    0.83
    etta
    0.83
    Activations Density 0.113%

    No Known Activations