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. 3-RES-JB
    6. 6854
    Prev
    Next
    INDEX
    Explanations

    mentions of the word "gasoline" in the text

    oai_token-act-pair · gpt-3.5-turbo

    references to gasoline and petrol prices

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    arat
    -0.80
    èª
    -0.76
     Proud
    -0.72
    achu
    -0.72
    ivas
    -0.71
    osen
    -0.67
    orld
    -0.66
    lihood
    -0.66
    eal
    -0.64
     CLSID
    -0.64
    POSITIVE LOGITS
     gasoline
    1.15
    oline
    1.03
     fuel
    0.98
     fumes
    0.92
     consumption
    0.83
     pedal
    0.83
     petrol
    0.82
     fueling
    0.82
     piston
    0.82
     excise
    0.82
    Activations Density 0.010%

    No Known Activations