© 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. 4-RES-JB
    11. 8495
    Prev
    Next
    INDEX
    Explanations

    unexpected or intriguing details or information

    oai_token-act-pair · gpt-3.5-turbo

    the word "interestingly" and its variants, indicating a focus on statements of surprise or notable facts

    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
     newsletters
    -0.70
    utm
    -0.69
    "},"
    -0.67
     throats
    -0.65
    iens
    -0.63
    aden
    -0.63
    tein
    -0.63
    ingham
    -0.63
    aily
    -0.62
    needs
    -0.61
    POSITIVE LOGITS
    Interestingly
    0.71
    theless
    0.69
     Interestingly
    0.68
     interestingly
    0.68
    lihood
    0.68
     Ironically
    0.68
    NESS
    0.67
    ety
    0.66
     paradox
    0.66
     coinc
    0.65
    Activations Density 0.017%

    No Known Activations