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

    Neuronpedia

    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. 8-RES_POST_32K-OAI
    11. 5781
    Prev
    Next
    INDEX
    Explanations

    mentions of demand in various contexts

    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
    442
    +0.14
    0.7%
    540
    +0.11
    0.5%
    427
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    442
    +0.14
    0.03
    540
    +0.11
    0.03
    272
    +0.10
    0.02
    Negative Logits
    issance
    -0.75
    kered
    -0.72
    =-=-=-=-
    -0.71
    beit
    -0.64
    cester
    -0.64
    ned
    -0.61
    swick
    -0.60
    istical
    -0.60
    Ń·
    -0.60
    paio
    -0.60
    POSITIVE LOGITS
    igslist
    0.94
    ele
    0.80
    Lex
    0.76
    л
    0.71
     fulfillment
    0.68
    Item
    0.67
     cooker
    0.67
    oux
    0.65
    Animation
    0.64
    force
    0.63
    Activations Density 0.053%

    No Known Activations