© 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. 4718
    Prev
    Next
    INDEX
    Explanations

    instances of commas in text

    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
    373
    +0.11
    0.5%
    316
    +0.10
    0.5%
    680
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    512
    +0.11
    0.05
    680
    +0.10
    0.04
    212
    +0.10
    0.03
    Negative Logits
     lod
    -0.59
     vain
    -0.56
     Brach
    -0.56
     completion
    -0.56
     galleries
    -0.55
     bombshell
    -0.54
     slee
    -0.54
     long
    -0.54
    CLUD
    -0.54
     Bellev
    -0.54
    POSITIVE LOGITS
    024
    0.95
    800
    0.86
    920
    0.84
    200
    0.84
    600
    0.83
    400
    0.81
    700
    0.81
    237
    0.81
    atile
    0.79
    300
    0.76
    Activations Density 0.157%

    No Known Activations