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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 6612
    Prev
    Next
    INDEX
    Explanations

    references to policies and agreements

    oai_token-act-pair · gpt-3.5-turbo

    mentions of "policy."

    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
     shave
    -0.72
    phabet
    -0.66
     cruc
    -0.66
     inspir
    -0.66
    teenth
    -0.66
     cast
    -0.65
    lins
    -0.64
    Ò
    -0.64
    veyard
    -0.64
     lik
    -0.63
    POSITIVE LOGITS
     Policy
    3.74
    Policy
    2.81
     Policies
    2.65
    policy
    2.56
     policy
    2.31
     policies
    1.75
     POL
    1.41
     Agenda
    1.32
     Legislation
    1.30
     Strategy
    1.26
    Activations Density 0.010%

    No Known Activations