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. 713
    Prev
    Next
    INDEX
    Explanations

    the last name "Varoufakis" in the text

    oai_token-act-pair · gpt-3.5-turbo

    the name "akis," indicating a focus on a specific individual or entity

    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
     Dru
    -0.89
     scouting
    -0.75
     Wild
    -0.72
     Survivor
    -0.71
     Poison
    -0.70
    nikov
    -0.70
     patented
    -0.69
     Fantastic
    -0.67
    wild
    -0.67
     stockpile
    -0.67
    POSITIVE LOGITS
    akis
    3.67
     Mb
    1.33
    ilk
    1.24
    LB
    1.22
    aghan
    1.10
    aki
    1.03
    mb
    1.01
     LB
    1.00
    onge
    0.98
    MB
    0.92
    Activations Density 0.027%

    No Known Activations