© 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. 4-RES-JB
    11. 17288
    Prev
    Next
    INDEX
    Explanations

    words related to interaction, specifically the act of engaging with someone or something

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "engage" and its variations, indicating a focus on involvement or participation

    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
    ë
    -0.71
    etry
    -0.69
    bey
    -0.63
    pled
    -0.63
    patient
    -0.62
    runners
    -0.61
     Bam
    -0.61
     surviving
    -0.61
    ple
    -0.61
    ark
    -0.60
    POSITIVE LOGITS
    ments
    0.99
    TAIN
    0.85
    odox
    0.83
    ienced
    0.82
     engaged
    0.82
    ment
    0.76
     engagement
    0.76
    poke
    0.75
    lished
    0.73
    #$
    0.72
    Activations Density 0.012%

    No Known Activations