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. 0-RES-JB
    6. 4488
    Prev
    Next
    INDEX
    Explanations

    words related to suggesting or proposing ideas

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "suggesting" and its variants

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    issance
    -0.70
    __
    -0.62
    deal
    -0.62
    oir
    -0.59
    ime
    -0.58
     grind
    -0.58
    vae
    -0.57
    oped
    -0.57
    hene
    -0.57
    &
    -0.57
    POSITIVE LOGITS
     suggesting
    3.31
     implying
    2.59
     indicating
    2.38
     proposing
    1.79
     recommending
    1.78
     confirming
    1.75
    suggest
    1.73
     predicting
    1.71
     illustrating
    1.64
     insisting
    1.60
    Activations Density 0.016%

    No Known Activations