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. 20777
    Prev
    Next
    INDEX
    Explanations

    phrases with the conjunction 'and' along with an element of comparison

    oai_token-act-pair · gpt-3.5-turbo

    the conjunction "and" to identify connected phrases or ideas

    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
    cised
    -0.66
    UGE
    -0.66
     Powered
    -0.61
    sung
    -0.60
    atan
    -0.59
    Share
    -0.59
    lections
    -0.58
    owered
    -0.57
    utenberg
    -0.56
    Constructed
    -0.56
    POSITIVE LOGITS
     hence
    1.49
     thus
    1.42
     therefore
    1.41
     consequently
    1.38
     thereby
    1.28
     secondly
    1.21
     furthermore
    1.13
     moreover
    0.97
     THEN
    0.97
     vice
    0.93
    Activations Density 0.423%

    No Known Activations