© 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. 10-RES_POST_32K-OAI
    11. 29203
    Prev
    Next
    INDEX
    Explanations

    phrases related to copyright and licensing terms

    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
    373
    +0.13
    0.6%
    324
    +0.10
    0.5%
    46
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    284
    +0.13
    0.03
    158
    +0.10
    0.03
    711
    +0.10
    0.01
    Negative Logits
    ĸļ
    -0.81
     reperto
    -0.77
    essor
    -0.68
    icho
    -0.66
    ources
    -0.64
    builders
    -0.63
    entin
    -0.63
    arthed
    -0.61
    Begin
    -0.59
    ebus
    -0.59
    POSITIVE LOGITS
     damages
    0.85
     DAM
    0.74
    ellectual
    0.66
    aneous
    0.62
     WARRANT
    0.60
    Premium
    0.60
     warranties
    0.60
    PLIED
    0.59
     depreciation
    0.59
    isable
    0.59
    Activations Density 0.254%

    No Known Activations