© 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. 10-RES_POST_32K-OAI
    11. 29795
    Prev
    Next
    INDEX
    Explanations

    prepositions and conjunctions that indicate relationships between ideas or clauses

    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
    224
    +0.10
    0.4%
    311
    +0.09
    0.4%
    356
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    394
    +0.10
    0.01
    373
    +0.09
    -0.01
    356
    +0.09
    0.01
    Negative Logits
    /**
    -0.74
    ffe
    -0.70
    adata
    -0.70
    kefeller
    -0.65
    acons
    -0.64
     âľĶ
    -0.64
    anski
    -0.63
    geist
    -0.61
    itory
    -0.60
    aji
    -0.60
    POSITIVE LOGITS
    »Ĵ
    0.83
     Squirrel
    0.68
    addle
    0.68
     guiActiveUnfocused
    0.63
    leneck
    0.61
    weeney
    0.57
     Cel
    0.57
    igmat
    0.55
    MpServer
    0.54
     deform
    0.54
    Activations Density 0.023%

    No Known Activations