© 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. 8-RES_POST_32K-OAI
    11. 6992
    Prev
    Next
    INDEX
    Explanations

    variations of the suffix "ra" or "ri" in words

    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
    705
    +0.10
    0.5%
    320
    +0.10
    0.5%
    107
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.10
    0.04
    373
    +0.10
    -0.02
    404
    +0.10
    0.04
    Negative Logits
     NX
    -0.70
    aceous
    -0.59
     Cosponsors
    -0.58
     Doctrine
    -0.57
     resemb
    -0.55
    lar
    -0.54
    eland
    -0.53
    orian
    -0.53
     similarities
    -0.53
    OVER
    -0.52
    POSITIVE LOGITS
    vernment
    1.10
    ptions
    0.91
    pta
    0.89
    cipline
    0.84
    icing
    0.81
    brance
    0.81
    ption
    0.74
    pped
    0.73
    cipl
    0.73
    erc
    0.72
    Activations Density 0.245%

    No Known Activations