© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 10-RES_POST_32K-OAI
    11. 30865
    Prev
    Next
    INDEX
    Explanations

    phrases indicating the existence and status of companies and their activities over time

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_10.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    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
    679
    +0.13
    0.6%
    102
    +0.12
    0.6%
    572
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    102
    +0.13
    0.16
    55
    +0.12
    0.14
    566
    +0.09
    0.11
    Negative Logits
     sightings
    -0.75
    kees
    -0.72
     jog
    -0.70
    pered
    -0.70
    jad
    -0.70
     Temper
    -0.69
    ukong
    -0.69
    Scroll
    -0.69
     encounters
    -0.68
     interpreter
    -0.67
    POSITIVE LOGITS
     headquartered
    1.46
     profitable
    1.33
     owned
    1.22
     subsidiaries
    1.06
     sued
    1.04
     franch
    1.03
     bankrupt
    1.03
    owned
    1.00
     suing
    0.99
     sponsoring
    0.99
    Activations Density 0.893%

    No Known Activations