© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 26337
    Prev
    Next
    INDEX
    Explanations

    phrases that describe various types of actions, decisions, or labels related to political or ethical contexts

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Neuron Alignment
    Index
    Value
    % of L₁
    374
    +0.14
    0.7%
    456
    +0.12
    0.5%
    65
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    36
    +0.14
    0.11
    374
    +0.12
    0.01
    512
    +0.11
    0.07
    Negative Logits
    groups
    -0.69
    limits
    -0.68
    izons
    -0.66
    existence
    -0.65
    iencies
    -0.64
    existent
    -0.64
    eon
    -0.63
    iverse
    -0.62
    sheets
    -0.62
     XI
    -0.62
    POSITIVE LOGITS
     setback
    1.06
     milestone
    1.04
     watershed
    1.03
     victory
    1.01
     reminder
    1.00
     triumph
    1.00
     blow
    0.99
     harb
    0.97
     testament
    0.96
     rebuke
    0.96
    Activations Density 0.680%

    No Known Activations