© 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
    3. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 8-RES_POST_32K-OAI
    11. 7659
    Prev
    Next
    INDEX
    Explanations

    references to geographic locations and political entities

    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
    130
    +0.11
    0.5%
    483
    +0.10
    0.4%
    373
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    412
    +0.11
    0.02
    241
    +0.10
    0.02
    373
    +0.10
    -0.02
    Negative Logits
    ĵĺ
    -0.74
    forth
    -0.65
    osponsors
    -0.58
     metic
    -0.57
     wid
    -0.57
    croft
    -0.55
     streng
    -0.54
    rone
    -0.54
     confir
    -0.54
    nan
    -0.54
    POSITIVE LOGITS
     Lines
    0.82
    azo
    0.81
    ña
    0.80
    ãĥŁ
    0.76
    akeru
    0.75
    inho
    0.73
     Observatory
    0.71
    esan
    0.70
     Janeiro
    0.70
    aminer
    0.68
    Activations Density 0.109%

    No Known Activations