© 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. 28317
    Prev
    Next
    INDEX
    Explanations

    instances of the word "come" and its variations in different 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
    46
    +0.12
    0.5%
    185
    +0.11
    0.5%
    545
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    377
    +0.12
    0.04
    242
    +0.11
    0.02
    545
    +0.10
    0.02
    Negative Logits
    aires
    -0.70
    efe
    -0.68
    adish
    -0.68
     guiActiveUn
    -0.65
    nec
    -0.63
     constitu
    -0.62
    è¦ļéĨĴ
    -0.60
     Carnage
    -0.59
    famous
    -0.58
     Everest
    -0.58
    POSITIVE LOGITS
     sit
    0.98
     closer
    0.95
     nearer
    0.83
     lie
    0.80
     here
    0.79
    Sit
    0.76
     ahead
    0.75
     awhile
    0.75
     along
    0.73
     behind
    0.72
    Activations Density 0.164%

    No Known Activations