© 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. 18190
    Prev
    Next
    INDEX
    Explanations

    specific instances of the word "strings."

    oai_token-act-pair · gpt-3.5-turbo

    references to "strings" in various contexts

    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
     Ath
    -0.96
    juven
    -0.88
    ellectual
    -0.77
    uben
    -0.72
     Nau
    -0.67
     Scout
    -0.67
    DEN
    -0.66
    ake
    -0.65
     Rapt
    -0.65
    ha
    -0.63
    POSITIVE LOGITS
     strings
    4.05
    strings
    2.88
     string
    1.89
    string
    1.75
    String
    1.40
     levers
    1.32
     String
    1.30
     cords
    1.27
     ropes
    1.27
     keyboards
    1.26
    Activations Density 0.021%

    No Known Activations