© 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. 5783
    Prev
    Next
    INDEX
    Explanations

    instances of the word "let" and its variations

    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
    22
    +0.10
    0.5%
    538
    +0.10
    0.4%
    70
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    108
    +0.10
    0.04
    611
    +0.10
    0.04
    294
    +0.09
    0.03
    Negative Logits
    Interstitial
    -0.89
    bda
    -0.82
    å§
    -0.81
    eem
    -0.79
    acci
    -0.79
    å£
    -0.77
    ãĤ¦ãĤ¹
    -0.75
    estones
    -0.73
     <[
    -0.73
    ä¹ĭ
    -0.72
    POSITIVE LOGITS
     employees
    0.98
     shoppers
    0.97
     corporations
    0.96
     consumers
    0.94
     retailers
    0.91
     people
    0.91
     companies
    0.90
     folks
    0.89
     stakeholders
    0.88
     customers
    0.87
    Activations Density 0.184%

    No Known Activations