© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 0-RES-JB
    11. 1414
    Prev
    Next
    INDEX
    Explanations

    mentions of social media followers or disciples

    oai_token-act-pair · gpt-3.5-turbo

    mentions of social media followers

    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
    DOC
    -0.69
     ..............
    -0.68
     NAFTA
    -0.64
    inn
    -0.64
    inventoryQuantity
    -0.64
     vi
    -0.62
    sweet
    -0.62
    uncture
    -0.62
    azard
    -0.61
     Hazard
    -0.61
    POSITIVE LOGITS
     followers
    3.69
     follower
    2.60
     adherents
    2.33
     disciples
    1.99
     supporters
    1.85
    ollower
    1.71
     listeners
    1.61
     believers
    1.59
     subscribers
    1.53
     fans
    1.47
    Activations Density 0.016%

    No Known Activations