© 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. 10-RES_POST_32K-OAI
    11. 29959
    Prev
    Next
    INDEX
    Explanations

    temporal markers or time-related references

    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
    373
    +0.11
    0.5%
    718
    +0.09
    0.4%
    237
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    762
    +0.11
    0.06
    377
    +0.09
    0.05
    225
    +0.09
    0.04
    Negative Logits
    quickShipAvailable
    -0.84
    ãĥ¤
    -0.77
    ibles
    -0.74
    ariat
    -0.71
    oir
    -0.70
    Marginal
    -0.69
    resa
    -0.66
    anth
    -0.66
    ahime
    -0.66
    nos
    -0.66
    POSITIVE LOGITS
     throughout
    0.81
     depending
    0.77
     across
    0.73
     until
    0.72
    ciating
    0.72
     begging
    0.69
     alike
    0.69
    )=(
    0.63
     strugg
    0.61
     unsuccessfully
    0.61
    Activations Density 0.595%

    No Known Activations