© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 10-RES-JB
    6. 13763
    Prev
    Next
    INDEX
    Explanations

    phrases related to fighting, resisting, and taking action for a cause or belief

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to advocacy and fighting for causes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     glances
    -0.74
     additions
    -0.71
     averages
    -0.70
     averaging
    -0.67
     calculations
    -0.67
    comments
    -0.64
     elig
    -0.63
     applicants
    -0.63
    Lot
    -0.63
     flashes
    -0.62
    POSITIVE LOGITS
     defend
    1.61
     liberate
    1.50
     protect
    1.45
     overthrow
    1.42
     oppose
    1.39
     eradicate
    1.37
     uphold
    1.35
     preserve
    1.35
     topple
    1.32
     restore
    1.31
    Activations Density 0.246%

    No Known Activations