© 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. 10-RES_POST_32K-OAI
    6. 29968
    Prev
    Next
    INDEX
    Explanations

    references to food storage and handling instructions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_10.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    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
    678
    +0.12
    0.5%
    732
    +0.10
    0.4%
    682
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    678
    +0.12
    0.13
    595
    +0.10
    0.04
    289
    +0.09
    0.00
    Negative Logits
    senal
    -0.72
    otaur
    -0.70
     Translation
    -0.66
     genesis
    -0.66
    profits
    -0.66
    oppable
    -0.65
    reply
    -0.65
     Valhalla
    -0.63
     merger
    -0.63
    arat
    -0.63
    POSITIVE LOGITS
     disinfect
    1.07
     moist
    1.00
     towels
    1.00
     towel
    0.93
     preferably
    0.91
     refriger
    0.91
     sterile
    0.90
    washing
    0.89
     gloves
    0.88
     tightly
    0.87
    Activations Density 2.113%

    No Known Activations