© 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 EvalsExportsinterp-engineNEWAPI 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. 27489
    Prev
    Next
    INDEX
    Explanations

    instances of code or programming language syntax, particularly related to numerical computations or algorithms

    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
    +0.15
    0.7%
    326
    +0.11
    0.5%
    627
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.15
    0.04
    678
    +0.11
    0.03
    314
    +0.10
    -0.00
    Negative Logits
     whims
    -0.77
     agendas
    -0.69
     frivolous
    -0.69
     Monsanto
    -0.69
     USDA
    -0.66
     UNESCO
    -0.65
     unheard
    -0.64
     crooked
    -0.64
     Himal
    -0.64
     magn
    -0.63
    POSITIVE LOGITS
    Storage
    1.15
    Connection
    1.04
    vec
    1.02
    wrap
    1.02
    Memory
    0.97
    DEP
    0.96
    Array
    0.95
    Integ
    0.94
    Window
    0.94
    json
    0.92
    Activations Density 0.183%

    No Known Activations