© 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. 1529
    Prev
    Next
    INDEX
    Explanations

    personal pronouns (e.g., "his," "their") and verbs expressing emotions like "expressing" in relation to public figures

    oai_token-act-pair · gpt-3.5-turbo

    expressions of personal opinions and feelings

    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
    atel
    -0.74
     Pwr
    -0.73
    £ı
    -0.70
    batch
    -0.67
     Rout
    -0.67
    oldown
    -0.67
    adobe
    -0.66
    acles
    -0.65
    asus
    -0.65
    veyard
    -0.64
    POSITIVE LOGITS
     displeasure
    1.73
     disapproval
    1.46
     dissatisfaction
    1.43
     admiration
    1.35
     gratitude
    1.32
     intention
    1.31
     disgust
    1.31
     disappointment
    1.30
     desire
    1.30
     disdain
    1.30
    Activations Density 0.120%

    No Known Activations