Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 4-RES-JB
    6. 23446
    Prev
    Next
    INDEX
    Explanations

    numeric values within parentheses, potentially related to prices or quantities

    oai_token-act-pair · gpt-3.5-turbo

    the presence of parentheses in the text

    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
     therap
    -0.76
     millenn
    -0.75
     lull
    -0.75
     wildlife
    -0.74
     psychiat
    -0.73
     intensive
    -0.71
     nutrient
    -0.69
     star
    -0.68
     retard
    -0.68
     vigil
    -0.67
    POSITIVE LOGITS
    including
    1.43
    sic
    1.42
    possibly
    1.39
    usually
    1.37
    although
    1.37
    which
    1.36
    albeit
    1.34
    perhaps
    1.32
    especially
    1.32
    often
    1.29
    Activations Density 0.221%

    No Known Activations