© 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. 6791
    Prev
    Next
    INDEX
    Explanations

    numeric values, particularly those related to financial figures or quantities

    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
    373
    +0.13
    0.6%
    573
    +0.12
    0.5%
    512
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    573
    +0.13
    0.04
    165
    +0.12
    0.04
    495
    +0.12
    0.05
    Negative Logits
    heres
    -0.67
     noses
    -0.67
     tensions
    -0.66
     tension
    -0.65
     throats
    -0.63
     fug
    -0.62
     catast
    -0.61
     timet
    -0.60
    ierrez
    -0.59
     tending
    -0.59
    POSITIVE LOGITS
    250
    0.85
    125
    0.81
    2500
    0.78
    625
    0.78
    ŃĶ
    0.76
    5000
    0.76
     trillion
    0.74
    billion
    0.74
    750
    0.74
    /+
    0.73
    Activations Density 0.256%

    No Known Activations