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. 1-RES-JB
    6. 8379
    Prev
    Next
    INDEX
    Explanations

    calls to action phrases

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to subscription or sign-up prompts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Fidel
    -0.71
    >]
    -0.68
     den
    -0.65
    _.
    -0.62
    士
    -0.62
    æĪ¦
    -0.62
     sting
    -0.62
     num
    -0.61
    minecraft
    -0.61
    ________________________
    -0.61
    POSITIVE LOGITS
    stairs
    1.33
    grades
    1.19
    dates
    1.10
    graded
    1.10
    dating
    1.09
    grading
    1.06
    rights
    1.00
    olicy
    0.96
    adesh
    0.92
    river
    0.91
    Activations Density 0.014%

    No Known Activations