© 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. 11-RES-JB
    11. 4134
    Prev
    Next
    INDEX
    Explanations

    ideologies or beliefs related to different "-isms."

    oai_token-act-pair · gpt-3.5-turbo

    variations and discussions related to different political ideologies, particularly those ending in 'ism'

    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
    umm
    -0.73
    fman
    -0.71
    liest
    -0.64
    âĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢ
    -0.63
     Chiefs
    -0.61
     Madden
    -0.60
     pad
    -0.59
     Force
    -0.59
    force
    -0.59
     Rams
    -0.59
    POSITIVE LOGITS
    yip
    1.17
    atism
    1.03
    ophobic
    1.02
    ophobia
    1.00
    ologies
    0.95
    ism
    0.92
    ophob
    0.90
    icist
    0.86
    edia
    0.86
    anism
    0.83
    Activations Density 0.171%

    No Known Activations