© 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. 6886
    Prev
    Next
    INDEX
    Explanations

    phrases related to reducing or minimizing unwanted effects or 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
    500
    +0.12
    0.6%
    292
    +0.10
    0.5%
    699
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.12
    0.04
    636
    +0.10
    0.03
    660
    +0.10
    0.03
    Negative Logits
    rings
    -0.71
    haps
    -0.69
    ouse
    -0.62
    mire
    -0.62
    rown
    -0.61
    icion
    -0.60
    ADRA
    -0.60
     largeDownload
    -0.58
    itals
    -0.58
    ĸļ
    -0.57
    POSITIVE LOGITS
     reliance
    1.01
     incidence
    0.98
     friction
    0.93
     footprint
    0.91
     impact
    0.91
     effectiveness
    0.90
     likelihood
    0.89
     dependence
    0.84
     unwanted
    0.83
     amount
    0.82
    Activations Density 0.244%

    No Known Activations