© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 4837
    Prev
    Next
    INDEX
    Explanations

    words related to maintaining or supporting individuals and communities

    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
    719
    +0.10
    0.4%
    129
    +0.09
    0.4%
    7
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    435
    +0.10
    0.03
    531
    +0.09
    0.03
    359
    +0.08
    0.03
    Negative Logits
    etheless
    -0.73
    ashi
    -0.72
     Shutterstock
    -0.70
    should
    -0.69
    haps
    -0.67
    might
    -0.66
     Maher
    -0.64
    assetsadobe
    -0.64
     Billboard
    -0.61
    :{
    -0.61
    POSITIVE LOGITS
     alive
    1.11
     afloat
    1.00
     accountable
    0.86
     happy
    0.86
     entertained
    0.84
     appealing
    0.83
     honest
    0.83
     solvent
    0.80
     guessing
    0.80
     safe
    0.76
    Activations Density 0.202%

    No Known Activations