© 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. 1-RES-JB
    11. 13129
    Prev
    Next
    INDEX
    Explanations

    the word "serpent" or variations of the word

    oai_token-act-pair · gpt-3.5-turbo

    references to "ser" as a word and related terms, possibly indicating a focus on servitude or service roles

    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
    ORE
    -0.86
    ORED
    -0.79
    Known
    -0.79
    INGTON
    -0.78
    hetti
    -0.78
    ORN
    -0.73
    OWS
    -0.72
    Pope
    -0.69
    ERC
    -0.69
    STEP
    -0.69
    POSITIVE LOGITS
     ser
    1.13
     confir
    1.11
    geant
    1.00
     exha
    0.94
     encount
    0.88
    ials
    0.84
     tyr
    0.83
     ende
    0.83
     lapt
    0.81
    itage
    0.80
    Activations Density 0.004%

    No Known Activations