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. 11-RES-JB
    6. 16199
    Prev
    Next
    INDEX
    Explanations

    numeric quantities and specifications within a text

    oai_token-act-pair · gpt-3.5-turbo

    numeric values and frequencies related to data structures or algorithms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    prise
    -0.79
    ulture
    -0.78
    vention
    -0.76
    urban
    -0.74
    gow
    -0.73
    section
    -0.72
     Charity
    -0.71
    awa
    -0.71
    ibus
    -0.70
     Fargo
    -0.70
    POSITIVE LOGITS
     atoms
    1.45
     entries
    1.31
     occurrences
    1.29
     inputs
    1.28
     bytes
    1.28
     digits
    1.28
     rows
    1.27
     integers
    1.27
     objects
    1.26
     elements
    1.26
    Activations Density 0.279%

    No Known Activations