Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 8-RES-JB
    6. 17528
    Prev
    Next
    INDEX
    Explanations

    terms related to socio-political and economic concepts, particularly focusing on middle class, foundational concepts, political manifestos, elites, populism, and real estate

    oai_token-act-pair · gpt-3.5-turbo

    terms related to socio-economic classes and political ideologies

    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
     Helpful
    -0.75
    idious
    -0.70
    inarily
    -0.66
    eous
    -0.66
     Kind
    -0.65
    rely
    -0.65
     embodiments
    -0.62
     Wicked
    -0.62
     Curious
    -0.61
    Houston
    -0.60
    POSITIVE LOGITS
    ="/
    0.81
    lance
    0.79
    archive
    0.76
    atorium
    0.73
    arist
    0.71
     ambassador
    0.71
     machine
    0.69
    mbudsman
    0.69
    scill
    0.68
    sword
    0.68
    Activations Density 0.364%

    No Known Activations