© 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 EvalsExportsinterp-engineNEWAPI 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. 8433
    Prev
    Next
    INDEX
    Explanations

    components of names or titles, particularly those that are repeated or form patterns

    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
    241
    +0.11
    0.5%
    20
    +0.10
    0.4%
    702
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.11
    0.03
    406
    +0.10
    0.02
    701
    +0.10
    0.02
    Negative Logits
    ©¶æ
    -0.74
    ãĤ©
    -0.69
    afety
    -0.67
    FTWARE
    -0.60
    ocument
    -0.58
     Supplemental
    -0.55
    urnal
    -0.55
    userc
    -0.55
    olves
    -0.54
     Surviv
    -0.54
    POSITIVE LOGITS
    agos
    0.89
    gow
    0.81
    azeera
    0.76
    hea
    0.74
    quin
    0.73
    adr
    0.72
    furt
    0.71
    reath
    0.70
    rant
    0.70
    oha
    0.69
    Activations Density 0.113%

    No Known Activations