© 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. 0-RES-JB
    11. 6778
    Prev
    Next
    INDEX
    Explanations

    words related to justice and legal terms

    oai_token-act-pair · gpt-3.5-turbo

    terms related to justice and social justice issues

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     residual
    -0.66
     dummy
    -0.66
     hat
    -0.61
    Introduced
    -0.59
     facial
    -0.59
     boil
    -0.59
     independently
    -0.58
     ground
    -0.58
    Tip
    -0.58
     Raspberry
    -0.57
    POSITIVE LOGITS
    justice
    3.97
    juries
    2.00
    jury
    1.84
    equality
    1.73
    merce
    1.29
    oha
    1.29
    truth
    1.18
    adequ
    1.16
    jured
    1.12
    reality
    1.01
    Activations Density 0.029%

    No Known Activations