© 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. 7089
    Prev
    Next
    INDEX
    Explanations

    references to specific titles and their connections to individuals or concepts

    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
    70
    +0.13
    0.6%
    663
    +0.10
    0.5%
    306
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    70
    +0.13
    0.10
    617
    +0.10
    0.04
    663
    +0.10
    0.02
    Negative Logits
    ifference
    -0.73
    artifacts
    -0.70
     imped
    -0.66
    gression
    -0.65
    ocent
    -0.61
    ographic
    -0.59
    ictive
    -0.59
     threads
    -0.59
    cember
    -0.59
    udeau
    -0.58
    POSITIVE LOGITS
     acronym
    1.43
     moniker
    1.36
     coined
    1.28
     nickname
    1.26
     name
    1.24
     abbre
    1.24
     slang
    1.22
     term
    1.19
     acron
    1.10
     pseudonym
    1.09
    Activations Density 4.654%

    No Known Activations