© 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 EvalsExportsinterp-engineNEWAPI 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. 7-RES-JB
    11. 14450
    Prev
    Next
    INDEX
    Explanations

    references to writing or being involved in the act of writing

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "write" and its variations, indicating a focus on writing-related actions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 7-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    7
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Ĭ±
    -0.97
    agara
    -0.78
    EGA
    -0.72
    allows
    -0.71
     Unsure
    -0.68
    abe
    -0.66
     Ukrain
    -0.65
    azar
    -0.65
    ella
    -0.64
     Yel
    -0.62
    POSITIVE LOGITS
     poems
    0.90
    itatively
    0.87
    writing
    0.84
    acters
    0.83
     poetry
    0.83
    smanship
    0.82
     essays
    0.80
     journal
    0.79
     aloud
    0.79
    writer
    0.75
    Activations Density 0.050%

    No Known Activations