© 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. 8161
    Prev
    Next
    INDEX
    Explanations

    instances of phrases indicating purpose or intention

    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
    171
    +0.13
    0.6%
    337
    +0.12
    0.6%
    655
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    608
    +0.13
    0.12
    665
    +0.12
    0.10
    718
    +0.10
    0.10
    Negative Logits
     removal
    -0.70
     knots
    -0.67
     separation
    -0.66
     sued
    -0.65
     equivalents
    -0.64
     Prev
    -0.63
     suicide
    -0.63
     inconsist
    -0.62
     exceptions
    -0.61
     :(
    -0.60
    POSITIVE LOGITS
     learn
    1.01
     discuss
    1.01
    learn
    0.98
     reminis
    0.95
     educate
    0.89
     hone
    0.88
    cele
    0.87
     brainstorm
    0.86
     discussing
    0.85
     celebrate
    0.85
    Activations Density 0.867%

    No Known Activations