© 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 EvalsExportsAPI 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. 2-RES-JB
    11. 6470
    Prev
    Next
    INDEX
    Explanations

    references to being detained or confined against one's will

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ories
    -0.73
    orp
    -0.72
    amera
    -0.70
    laus
    -0.69
    rous
    -0.68
    ript
    -0.67
    opic
    -0.66
    alore
    -0.64
    orie
    -0.63
    gian
    -0.61
    POSITIVE LOGITS
     prisoners
    1.07
     inmates
    0.96
     prisoner
    0.95
     captives
    0.93
     detainees
    0.86
    Tube
    0.82
     inmate
    0.78
     confinement
    0.77
     Prison
    0.74
     hostage
    0.73
    Activations Density 0.010%

    No Known Activations