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. 18661
    Prev
    Next
    INDEX
    Explanations

    words related to the concept of fueling or being fueled

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to causes or motivations behind events or actions

    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
    perture
    -0.79
    roma
    -0.71
    ihad
    -0.70
    achu
    -0.69
    bestos
    -0.68
    opers
    -0.68
     WATCHED
    -0.66
    atar
    -0.66
    arat
    -0.65
    é¾įåĸļ士
    -0.64
    POSITIVE LOGITS
     fue
    1.06
     fueling
    1.00
     fueled
    0.96
     fuelled
    0.90
    Fuel
    0.83
    fuel
    0.79
    ItemTracker
    0.75
     fuel
    0.74
    plug
    0.74
    ument
    0.74
    Activations Density 0.015%

    No Known Activations