© 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. 11-RES-JB
    11. 7642
    Prev
    Next
    INDEX
    Explanations

    high-ranking job titles and positions

    oai_token-act-pair · gpt-3.5-turbo

    titles and roles related to executives and management positions

    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
    inventoryQuantity
    -0.81
    ulz
    -0.76
     Pastebin
    -0.75
    bags
    -0.74
    ertodd
    -0.74
    NetMessage
    -0.70
    elight
    -0.69
    olds
    -0.68
     Soldiers
    -0.68
    Skill
    -0.68
    POSITIVE LOGITS
     director
    1.59
     editor
    1.43
     vice
    1.33
     coordinator
    1.33
     curator
    1.27
     advisor
    1.25
     counsel
    1.23
    director
    1.21
     adviser
    1.21
     planner
    1.20
    Activations Density 0.077%

    No Known Activations