© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 8-RES-JB
    11. 927
    Prev
    Next
    INDEX
    Explanations

    the word "grace" or variations of it

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "grace" and its variations within various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    algia
    -0.72
     proprietary
    -0.62
    ramid
    -0.62
    rim
    -0.62
     hobby
    -0.61
    ocus
    -0.61
     Argon
    -0.59
    rogen
    -0.59
    riber
    -0.59
     ID
    -0.58
    POSITIVE LOGITS
    fully
    1.65
    ful
    1.35
    fulness
    1.32
    FUL
    1.10
    full
    1.03
    ibility
    0.95
    ifully
    0.89
    combe
    0.84
    ously
    0.79
    eous
    0.79
    Activations Density 0.044%

    No Known Activations