© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 1-RES-JB
    6. 13269
    Prev
    Next
    INDEX
    Explanations

    terms related to accidents or unintentional actions

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "accidentally" and related terms indicating unintentional actions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    leaders
    -0.73
    dar
    -0.72
     options
    -0.71
    nosis
    -0.71
    aily
    -0.71
     Techniques
    -0.70
    shall
    -0.70
     architectures
    -0.70
    states
    -0.67
     attorneys
    -0.67
    POSITIVE LOGITS
     accidentally
    1.33
     inadvertently
    1.04
     unintentionally
    1.03
     accidental
    0.93
     bumped
    0.87
     drown
    0.87
     spilled
    0.85
     contam
    0.84
     spilling
    0.82
     unwittingly
    0.82
    Activations Density 0.009%

    No Known Activations