© 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. 6952
    Prev
    Next
    INDEX
    Explanations

    references to the word "Kill" with varying contexts such as movie titles, character names, and actions

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "Kill" in various contexts

    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
    BUS
    -0.71
    umn
    -0.68
    ĸļ
    -0.68
    éĹ
    -0.66
    rial
    -0.65
    phrine
    -0.65
    issance
    -0.65
    KEN
    -0.64
    acle
    -0.63
     understanding
    -0.63
    POSITIVE LOGITS
    ings
    0.98
    joy
    0.95
    killer
    0.91
    ingly
    0.90
    adelphia
    0.89
    houses
    0.87
    guards
    0.87
    ishers
    0.86
    mails
    0.85
    boats
    0.84
    Activations Density 0.017%

    No Known Activations