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

    Neuronpedia

    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. 29730
    Prev
    Next
    INDEX
    Explanations

    elements related to mathematical or logical structures

    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
    314
    +0.14
    0.6%
    345
    +0.10
    0.4%
    182
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    96
    +0.14
    0.02
    6
    +0.10
    0.03
    275
    +0.09
    0.01
    Negative Logits
    achev
    -0.82
    ocating
    -0.72
     tuna
    -0.71
     overboard
    -0.70
     SAS
    -0.68
    icka
    -0.67
    paces
    -0.67
     throughput
    -0.67
     embargo
    -0.67
    poke
    -0.66
    POSITIVE LOGITS
    )}
    0.77
     Shinji
    0.74
     @@
    0.73
    ,,
    0.73
    Optional
    0.71
    Answer
    0.70
    Eva
    0.70
    Values
    0.68
    Ir
    0.68
    ãĢį
    0.68
    Activations Density 0.175%

    No Known Activations