© 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
    2. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    3. GPT2-Small
    4. Residual Post - 32k
    5. 8-RES_POST_32K-OAI
    6. 5647
    Prev
    Next
    INDEX
    Explanations

    intensifiers and descriptive adjectives indicating a high level of quality or degree

    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
    455
    +0.11
    0.5%
    120
    +0.10
    0.5%
    101
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    540
    +0.11
    0.09
    102
    +0.10
    0.08
    55
    +0.10
    0.08
    Negative Logits
    ibur
    -0.76
     Americans
    -0.70
    Correction
    -0.70
     Rodham
    -0.70
    terday
    -0.70
    Saharan
    -0.69
    things
    -0.69
    culosis
    -0.68
    UGH
    -0.68
     Doctors
    -0.67
    POSITIVE LOGITS
     versatile
    1.25
     customizable
    1.05
     forgiving
    0.97
     inexpensive
    0.95
     flexible
    0.94
     beginner
    0.91
     scalable
    0.88
     economical
    0.88
     efficient
    0.88
     portable
    0.88
    Activations Density 0.503%

    No Known Activations