© 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. 8519
    Prev
    Next
    INDEX
    Explanations

    the word "up" used in various 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
    125
    +0.13
    0.6%
    314
    +0.12
    0.5%
    99
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    125
    +0.13
    0.06
    262
    +0.12
    0.04
    274
    +0.11
    0.05
    Negative Logits
     Abyss
    -0.61
    999
    -0.59
    Anne
    -0.57
    9999
    -0.57
    ====
    -0.56
     CLSID
    -0.56
    mia
    -0.55
    izable
    -0.54
    mite
    -0.53
    othy
    -0.53
    POSITIVE LOGITS
    stairs
    1.03
    dates
    1.00
    rights
    0.98
    raised
    0.96
    grades
    0.94
    skirts
    0.92
    grading
    0.91
    river
    0.91
    graded
    0.85
    dating
    0.85
    Activations Density 0.344%

    No Known Activations