© 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. 5874
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "more"

    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
    68
    +0.11
    0.5%
    123
    +0.10
    0.5%
    402
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    402
    +0.11
    0.07
    726
    +0.10
    0.06
    408
    +0.09
    0.07
    Negative Logits
    ¬
    -0.91
    ħĭ
    -0.91
    ĪĴ
    -0.86
    ²
    -0.83
    ®
    -0.81
    ĸļ
    -0.78
    ocrat
    -0.76
    iggurat
    -0.75
    ģ
    -0.74
    º
    -0.73
    POSITIVE LOGITS
     entries
    1.20
     copies
    1.17
     instances
    1.14
     pieces
    1.13
     bottles
    1.06
     letters
    1.06
     units
    1.06
     submissions
    1.02
     items
    1.01
     batches
    1.01
    Activations Density 0.221%

    No Known Activations