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. 15613
    Prev
    Next
    INDEX
    Explanations

    the word "certain" with high activation values, indicating a focus on specific or particular instances

    oai_token-act-pair · gpt-3.5-turbo

    mentions of the word "certain."

    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
     thence
    -0.87
    ĺħ
    -0.81
    dl
    -0.77
    DragonMagazine
    -0.70
    ÅĤ
    -0.68
     Cum
    -0.67
    here
    -0.67
    Fal
    -0.65
    aline
    -0.65
    tip
    -0.65
    POSITIVE LOGITS
    ties
    1.62
     kinds
    1.40
     types
    1.30
     aspects
    1.22
     wavelengths
    1.06
     segments
    1.00
     sections
    0.98
     embodiments
    0.97
     parts
    0.97
     combinations
    0.97
    Activations Density 0.031%

    No Known Activations