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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 8-RES-JB
    11. 18608
    Prev
    Next
    INDEX
    Explanations

    phrases related to authority or directives from a higher power

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to political manipulation or direction

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     contrasts
    -0.73
    elaide
    -0.69
    mopolitan
    -0.67
    float
    -0.66
    Studies
    -0.65
     fossils
    -0.64
    izons
    -0.63
    Decre
    -0.63
     overfl
    -0.63
     wildfires
    -0.62
    POSITIVE LOGITS
     superiors
    1.13
     management
    1.12
     supervisors
    1.11
     supervisor
    1.08
     administrators
    1.05
     manag
    1.05
     admins
    1.03
     overse
    1.02
     managers
    1.01
     administrator
    1.00
    Activations Density 1.189%

    No Known Activations