© 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
    3. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 8-RES_POST_32K-OAI
    11. 4562
    Prev
    Next
    INDEX
    Explanations

    phrases that describe methods or approaches to achieving something

    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
    373
    +0.15
    0.7%
    386
    +0.11
    0.5%
    374
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    50
    +0.15
    0.04
    705
    +0.11
    0.04
    13
    +0.10
    0.03
    Negative Logits
     Names
    -0.71
    ibrary
    -0.68
    asions
    -0.68
     Fault
    -0.66
     Subst
    -0.62
    ancies
    -0.62
    ichen
    -0.61
    sbm
    -0.61
    entanyl
    -0.60
    tis
    -0.59
    POSITIVE LOGITS
    ãĤī
    0.67
    ÙĨ
    0.67
    EGA
    0.64
     showcase
    0.64
     continuation
    0.64
     homage
    0.64
     beginners
    0.63
     legitim
    0.62
     highlight
    0.62
    ILLE
    0.61
    Activations Density 0.182%

    No Known Activations