© 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. 8509
    Prev
    Next
    INDEX
    Explanations

    active voice constructions in various tenses

    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
    532
    +0.13
    0.6%
    55
    +0.12
    0.5%
    713
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    55
    +0.13
    0.09
    102
    +0.12
    0.09
    566
    +0.09
    0.07
    Negative Logits
     understandably
    -0.79
    entary
    -0.67
     clustered
    -0.63
    mentioned
    -0.61
     predictably
    -0.60
     evidently
    -0.59
     beautifully
    -0.59
     poster
    -0.58
     vocabulary
    -0.58
     seem
    -0.58
    POSITIVE LOGITS
     quitting
    0.81
     resign
    0.80
     PsyNet
    0.77
     intention
    0.74
     suspending
    0.69
     Accessed
    0.68
     sorry
    0.68
    aye
    0.68
    rael
    0.68
    iss
    0.67
    Activations Density 0.269%

    No Known Activations