© 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. 6434
    Prev
    Next
    INDEX
    Explanations

    repeated instances of the letter 'w' in a variety of contexts

    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
    496
    +0.13
    0.6%
    7
    +0.11
    0.5%
    360
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    99
    +0.13
    0.04
    438
    +0.11
    0.03
    18
    +0.10
    0.03
    Negative Logits
    terday
    -0.97
    Ô
    -0.95
    etheless
    -0.81
    ntil
    -0.69
     Lumpur
    -0.67
    INGTON
    -0.66
     Spectrum
    -0.66
    )=(
    -0.65
     Ended
    -0.64
    shire
    -0.64
    POSITIVE LOGITS
    oths
    0.89
    ester
    0.84
    ags
    0.81
    agers
    0.81
    imp
    0.80
    ogly
    0.79
    erning
    0.79
    advertising
    0.79
    evil
    0.78
    ild
    0.78
    Activations Density 0.046%

    No Known Activations