© 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. 30888
    Prev
    Next
    INDEX
    Explanations

    significant negative events or situations

    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
    42
    +0.11
    0.5%
    469
    +0.10
    0.5%
    206
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    241
    +0.11
    0.03
    193
    +0.10
    0.03
    679
    +0.10
    0.07
    Negative Logits
    urring
    -0.66
     mishand
    -0.64
    anship
    -0.63
    Series
    -0.63
    irez
    -0.62
    iple
    -0.62
    agging
    -0.60
    gency
    -0.60
    uyomi
    -0.59
    Ultra
    -0.59
    POSITIVE LOGITS
     here
    1.83
    here
    1.55
     Here
    1.53
     HERE
    1.46
    Here
    1.36
     hereafter
    1.02
     hither
    0.99
     herein
    0.94
     elsewhere
    0.93
    heres
    0.88
    Activations Density 1.182%

    No Known Activations