© 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. 0-RES-JB
    11. 15859
    Prev
    Next
    INDEX
    Explanations

    phrases related to beginning or initiating actions

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "started"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Deliver
    -0.63
    versely
    -0.60
    jury
    -0.60
     fidelity
    -0.59
     virtue
    -0.57
    margin
    -0.57
     representations
    -0.56
     persuasion
    -0.56
     judgment
    -0.55
    bern
    -0.55
    POSITIVE LOGITS
     started
    3.22
    started
    2.51
     Started
    2.51
     began
    2.48
     begun
    2.25
     commenced
    2.10
     resumed
    1.72
     starts
    1.71
     launched
    1.69
     start
    1.60
    Activations Density 0.026%

    No Known Activations