© 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
    3. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 8-RES_POST_32K-OAI
    11. 6501
    Prev
    Next
    INDEX
    Explanations

    elements related to programming or coding syntax, particularly assignments and function definitions

    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
    459
    +0.13
    0.6%
    373
    +0.12
    0.6%
    731
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    160
    +0.13
    0.07
    314
    +0.12
    -0.00
    284
    +0.08
    0.04
    Negative Logits
    orsi
    -0.77
     jails
    -0.72
    issance
    -0.68
    eworthy
    -0.67
     Pigs
    -0.66
     toll
    -0.62
     groceries
    -0.62
    lication
    -0.61
     galleries
    -0.60
    llo
    -0.59
    POSITIVE LOGITS
    utf
    1.01
     srfAttach
    0.82
    Adapt
    0.80
     Fug
    0.78
    Mini
    0.76
    Static
    0.74
    ãĤ¨
    0.73
    sharp
    0.72
    poly
    0.72
    ãĥķãĤ©
    0.72
    Activations Density 0.575%

    No Known Activations