© 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. 30004
    Prev
    Next
    INDEX
    Explanations

    occurrences of the name "Bob"

    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
    314
    +0.13
    0.6%
    223
    +0.11
    0.5%
    392
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.13
    0.01
    232
    +0.11
    0.01
    569
    +0.10
    0.01
    Negative Logits
    Reloaded
    -0.71
     Sunny
    -0.60
     Machina
    -0.58
    ashtra
    -0.58
     vetting
    -0.57
     embodiments
    -0.56
     Entity
    -0.55
     Duty
    -0.55
     plural
    -0.55
     Pavilion
    -0.55
    POSITIVE LOGITS
    bie
    1.37
    bi
    1.21
    bies
    1.16
    cats
    1.13
    cat
    1.10
    bing
    1.06
    ble
    1.06
    sle
    1.04
    bles
    1.02
    bers
    1.00
    Activations Density 0.045%

    No Known Activations