© 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. 8017
    Prev
    Next
    INDEX
    Explanations

    instances of the prefix "com," indicating a focus on terms related to combination or community

    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
    314
    +0.17
    0.8%
    746
    +0.10
    0.5%
    167
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    174
    +0.17
    0.03
    477
    +0.10
    0.02
    586
    +0.09
    0.02
    Negative Logits
     rave
    -0.63
     caution
    -0.62
     encl
    -0.62
     gratification
    -0.62
     Refuge
    -0.62
     recall
    -0.61
     fright
    -0.60
     steroids
    -0.59
     Wolves
    -0.59
     firsthand
    -0.58
    POSITIVE LOGITS
    puters
    1.37
    pleted
    1.35
    fortable
    1.34
    puting
    1.33
    pletion
    1.22
    pleting
    1.20
    parable
    1.15
    plement
    1.12
    edy
    1.11
    puted
    1.11
    Activations Density 0.072%

    No Known Activations