© 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. 8510
    Prev
    Next
    INDEX
    Explanations

    relationships and transitions involving the concept of succession or replacement

    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
    176
    +0.11
    0.5%
    420
    +0.10
    0.4%
    627
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    608
    +0.11
    0.14
    362
    +0.10
    0.12
    479
    +0.10
    0.07
    Negative Logits
    izontal
    -0.80
    utterstock
    -0.80
     reuse
    -0.78
    artifacts
    -0.77
    £ı
    -0.76
    ahime
    -0.75
    isphere
    -0.74
    nex
    -0.71
    EStreamFrame
    -0.70
     indic
    -0.70
    POSITIVE LOGITS
     George
    1.27
     Barack
    1.26
     Richard
    1.22
     Ronald
    1.20
     Nicolas
    1.19
     Adolf
    1.18
     John
    1.17
     Anthony
    1.17
     Osama
    1.17
     Lionel
    1.17
    Activations Density 1.939%

    No Known Activations