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. 14615
    Prev
    Next
    INDEX
    Explanations

    words related to different styles, such as artistic or fashion styles

    oai_token-act-pair · gpt-3.5-turbo

    references to various styles in general contexts

    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
     Investor
    -0.77
    ridor
    -0.66
    amine
    -0.65
     whistleblower
    -0.64
    mur
    -0.63
     operator
    -0.62
    ylum
    -0.62
    isition
    -0.62
    invest
    -0.61
    enary
    -0.61
    POSITIVE LOGITS
     styles
    3.46
    styles
    2.55
     Styles
    2.49
     genres
    1.92
     formats
    1.50
     sizes
    1.45
     styling
    1.40
    Sty
    1.40
     shapes
    1.35
     genders
    1.34
    Activations Density 0.027%

    No Known Activations