© 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. 8-RES_POST_32K-OAI
    11. 4563
    Prev
    Next
    INDEX
    Explanations

    recurring expressions or phrases within the text

    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
    710
    +0.13
    0.6%
    187
    +0.11
    0.5%
    526
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    644
    +0.13
    0.02
    725
    +0.11
    0.01
    274
    +0.10
    0.01
    Negative Logits
    fman
    -0.70
    ÄŁ
    -0.68
    gur
    -0.67
     Thro
    -0.67
    DERR
    -0.66
     Federation
    -0.64
     hemor
    -0.62
     Indies
    -0.61
     Tir
    -0.61
    jri
    -0.60
    POSITIVE LOGITS
    ology
    1.19
    phrase
    1.02
    witz
    0.91
    terday
    0.90
    ological
    0.88
    atre
    0.87
    stick
    0.86
    stress
    0.84
    obi
    0.83
    eting
    0.82
    Activations Density 0.066%

    No Known Activations