© 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. 30186
    Prev
    Next
    INDEX
    Explanations

    adjectives and adverbial phrases that convey strong judgments or characteristics

    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
    378
    +0.10
    0.5%
    722
    +0.10
    0.4%
    464
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.10
    0.13
    317
    +0.10
    0.07
    711
    +0.09
    0.02
    Negative Logits
     FANT
    -0.61
    conservancy
    -0.59
     fateful
    -0.58
    anders
    -0.57
    ournal
    -0.54
    ocations
    -0.54
    ODUCT
    -0.53
     GEAR
    -0.52
    angs
    -0.52
    ioxide
    -0.52
    POSITIVE LOGITS
    ly
    3.79
    LY
    2.59
    lys
    1.74
    fully
    1.67
    liness
    1.64
    lly
    1.58
    edly
    1.55
    liest
    1.54
    ELY
    1.50
    ially
    1.49
    Activations Density 2.720%

    No Known Activations