© 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. 14956
    Prev
    Next
    INDEX
    Explanations

    instances indicating various activities or items, often related to work or productivity

    oai_token-act-pair · gpt-3.5-turbo

    punctuation marks that signify pauses or separations in the text

    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
    orem
    -0.81
    xb
    -0.73
    emaker
    -0.73
    ãĥ£
    -0.71
    asma
    -0.71
    rio
    -0.71
    ropolis
    -0.70
    atorium
    -0.70
    xe
    -0.69
    oret
    -0.69
    POSITIVE LOGITS
     respectively
    2.19
     alike
    1.34
    depending
    1.17
     etc
    1.10
     depending
    1.05
     moreover
    1.00
    etc
    1.00
     whichever
    0.93
     among
    0.93
     plus
    0.89
    Activations Density 0.364%

    No Known Activations