© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 8-RES-JB
    11. 1251
    Prev
    Next
    INDEX
    Explanations

    phrases or sentences containing the words "take up" with an emphasis on actions or tasks being assumed or undertaken

    oai_token-act-pair · gpt-3.5-turbo

    phrases that include the concept of "taking up space" or "taking up room."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    lua
    -0.68
    hof
    -0.67
    come
    -0.64
    jong
    -0.64
    ificent
    -0.60
    gets
    -0.59
    nov
    -0.59
    ga
    -0.59
    arily
    -0.59
    Rus
    -0.56
    POSITIVE LOGITS
     residence
    1.10
    rights
    0.83
     positions
    0.74
     position
    0.73
     arms
    0.69
     stances
    0.68
    stantial
    0.67
     sugg
    0.66
    SpaceEngineers
    0.65
    ibilities
    0.64
    Activations Density 0.031%

    No Known Activations