© 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. 5298
    Prev
    Next
    INDEX
    Explanations

    terms related to environmental issues and policies

    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
    644
    +0.12
    0.6%
    634
    +0.12
    0.5%
    575
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    634
    +0.12
    0.03
    561
    +0.12
    0.03
    575
    +0.10
    0.03
    Negative Logits
    doms
    -0.80
    dom
    -0.78
    culus
    -0.76
    akia
    -0.68
    strap
    -0.67
    ework
    -0.66
    loading
    -0.64
    oons
    -0.61
    inal
    -0.60
    table
    -0.60
    POSITIVE LOGITS
     stewards
    0.85
    ist
    0.85
     hazards
    0.83
     degradation
    0.81
     pollut
    0.80
    ists
    0.79
     cleanup
    0.78
    conservancy
    0.77
     pollution
    0.77
     footprint
    0.76
    Activations Density 0.095%

    No Known Activations