© 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. 30444
    Prev
    Next
    INDEX
    Explanations

    negative aspects or criticisms, particularly in the context of research citations and 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
    547
    +0.12
    0.5%
    715
    +0.10
    0.5%
    378
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    96
    +0.12
    0.03
    18
    +0.10
    0.04
    430
    +0.10
    0.04
    Negative Logits
    retty
    -0.71
     autos
    -0.70
     sher
    -0.69
     lett
    -0.66
     Gou
    -0.66
    arya
    -0.65
    roth
    -0.64
    atha
    -0.63
    HAM
    -0.63
    athing
    -0.62
    POSITIVE LOGITS
     197
    1.09
    197
    1.09
     198
    1.07
    198
    1.04
     199
    0.94
     196
    0.91
    203
    0.88
    199
    0.88
    1972
    0.86
    202
    0.84
    Activations Density 0.266%

    No Known Activations