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. 11-RES-JB
    6. 24354
    Prev
    Next
    INDEX
    Explanations

    the word "game" with varying degrees of emphasis or descriptors

    oai_token-act-pair · gpt-3.5-turbo

    references to games and competitive scenarios

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    attery
    -0.79
     Osw
    -0.69
    ought
    -0.65
    onga
    -0.65
    aline
    -0.64
    atts
    -0.64
    von
    -0.63
     istg
    -0.63
    ayers
    -0.60
    allel
    -0.60
    POSITIVE LOGITS
     chang
    1.17
    plan
    1.12
     theoret
    1.09
    manship
    1.01
    keeper
    0.97
    plays
    0.96
    keepers
    0.94
    wright
    0.94
    play
    0.90
    cock
    0.89
    Activations Density 0.065%

    No Known Activations