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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 21713
    Prev
    Next
    INDEX
    Explanations

    phrases related to the name "Galvan."

    oai_token-act-pair · gpt-3.5-turbo

    repeated mentions of the name "Galvan."

    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
     mattress
    -0.77
     00000000
    -0.71
    vironment
    -0.67
     hypert
    -0.66
     __
    -0.63
     Nichols
    -0.63
     maple
    -0.63
     stump
    -0.61
    idays
    -0.61
     impeachment
    -0.61
    POSITIVE LOGITS
     Gal
    3.85
    Gal
    2.99
     gal
    1.82
    gal
    1.68
     Gall
    1.47
    Gall
    1.27
     Galactic
    1.18
     Gil
    1.12
     Bal
    1.11
     Ga
    1.08
    Activations Density 0.020%

    No Known Activations