© 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. 7704
    Prev
    Next
    INDEX
    Explanations

    specific references to reports and studies

    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
    219
    +0.14
    0.6%
    261
    +0.11
    0.5%
    749
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    655
    +0.14
    0.09
    679
    +0.11
    0.10
    139
    +0.10
    0.07
    Negative Logits
     downtime
    -0.69
     upkeep
    -0.68
     vanishing
    -0.62
     disappearing
    -0.61
    Cooldown
    -0.60
     srfAttach
    -0.60
    animate
    -0.59
    Offline
    -0.59
    bound
    -0.58
     VIDEOS
    -0.58
    POSITIVE LOGITS
    uggest
    1.25
     detailing
    1.10
     outlines
    1.07
     outlining
    1.06
     discusses
    0.98
     clar
    0.97
     summarize
    0.97
     summar
    0.94
     sheds
    0.94
     summarizes
    0.94
    Activations Density 0.874%

    No Known Activations