© 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. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 10-RES-JB
    11. 5778
    Prev
    Next
    INDEX
    Explanations

    the word "what" followed by a comparison or quantity in a sentence

    oai_token-act-pair · gpt-3.5-turbo

    references to comparisons or measurements against expectations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    artifacts
    -0.76
    nec
    -0.73
    lishing
    -0.73
    yi
    -0.65
    mad
    -0.63
    inning
    -0.62
    endix
    -0.61
    bold
    -0.61
    mile
    -0.60
    flag
    -0.60
    POSITIVE LOGITS
    soever
    0.92
     they
    0.90
     envisioned
    0.88
     happens
    0.88
     we
    0.84
     existed
    0.82
     historians
    0.81
     awaits
    0.80
     happened
    0.80
     mattered
    0.79
    Activations Density 0.097%

    No Known Activations