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

    Neuronpedia

    APICircuit 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. 8-RES-JB
    6. 21877
    Prev
    Next
    INDEX
    Explanations

    information related to numbers and stats

    oai_token-act-pair · gpt-3.5-turbo

    numerical data, specifically statistics and percentages

    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
     WARN
    -0.61
    tell
    -0.58
    onen
    -0.57
    reads
    -0.55
    SEE
    -0.53
    920
    -0.53
     Read
    -0.52
     Exit
    -0.52
     JO
    -0.52
    OSH
    -0.51
    POSITIVE LOGITS
    rador
    0.75
     TBA
    0.73
    auld
    0.72
     versa
    0.70
     apiece
    0.70
     undecided
    0.70
     respectively
    0.68
     finally
    0.68
    ogether
    0.67
    %).
    0.67
    Activations Density 0.265%

    No Known Activations