© 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. 30214
    Prev
    Next
    INDEX
    Explanations

    references to design quality and creativity

    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
    562
    +0.13
    0.6%
    196
    +0.11
    0.5%
    221
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    160
    +0.13
    0.28
    107
    +0.11
    0.27
    640
    +0.10
    0.19
    Negative Logits
    accompan
    -1.00
     advoc
    -0.84
     util
    -0.77
    osponsors
    -0.77
     Whilst
    -0.76
    ofi
    -0.76
    Introduction
    -0.75
     reconc
    -0.75
    ius
    -0.75
    ahime
    -0.74
    POSITIVE LOGITS
     Slate
    1.18
    Anyway
    1.16
    Advertisement
    1.14
     Nope
    1.12
     Gawker
    0.99
    Seriously
    0.92
     Yeah
    0.92
     anyway
    0.91
    Yeah
    0.90
     Weird
    0.88
    Activations Density 0.883%

    No Known Activations