© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    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. 10-RES_POST_32K-OAI
    11. 29011
    Prev
    Next
    INDEX
    Explanations

    instances of the word "soon" and related concepts of anticipation or progression in narrative contexts

    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
    38
    +0.12
    0.5%
    516
    +0.11
    0.5%
    54
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    139
    +0.12
    0.11
    551
    +0.11
    0.10
    718
    +0.10
    0.11
    Negative Logits
    yet
    -0.95
     tomorrow
    -0.75
    still
    -0.72
     NOW
    -0.71
    DAY
    -0.69
    cffffcc
    -0.67
     inaction
    -0.67
    currently
    -0.66
    PRES
    -0.66
     remains
    -0.65
    POSITIVE LOGITS
     realize
    0.88
     realise
    0.84
     realized
    0.82
     realizing
    0.80
     realised
    0.78
     become
    0.78
     realizes
    0.75
     loos
    0.75
     learned
    0.74
     morphed
    0.73
    Activations Density 1.968%

    No Known Activations