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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsBlogSlackPrivacy & 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. 24
    Prev
    Next
    INDEX
    Explanations

    words related to the chemical compound "hexane"

    oai_token-act-pair · gpt-3.5-turbo

    keywords related to chemical compounds and baking processes

    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
    Anderson
    -0.81
     Minority
    -0.79
     Surviv
    -0.75
     Anderson
    -0.74
     Till
    -0.74
    aver
    -0.72
     Limbaugh
    -0.72
    advertising
    -0.71
    IDER
    -0.69
     Burger
    -0.69
    POSITIVE LOGITS
     hex
    3.98
     Hex
    2.80
    hex
    2.67
     pent
    1.33
     oct
    1.25
     chrom
    1.19
     decimal
    1.09
     glyph
    1.04
     dracon
    0.95
     runes
    0.95
    Activations Density 0.035%

    No Known Activations