© 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. 4-RES-JB
    11. 1981
    Prev
    Next
    INDEX
    Explanations

    words or phrases related to the beginning or start of events or actions

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "start"

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

    No Comments

    Negative Logits
    phy
    -0.72
    illard
    -0.68
    iliary
    -0.65
    inarily
    -0.65
    ilitary
    -0.64
    eneg
    -0.64
    asca
    -0.63
     forensic
    -0.61
    hare
    -0.60
     bathrooms
    -0.60
    POSITIVE LOGITS
    nings
    1.26
    ners
    0.80
    watch
    0.77
    rek
    0.76
    ãĤ¿
    0.75
    ribune
    0.73
    starting
    0.73
     Reck
    0.72
    here
    0.68
    hend
    0.67
    Activations Density 0.028%

    No Known Activations