© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 16792
    Prev
    Next
    INDEX
    Explanations

    phrases related to disclosure of personal information

    oai_token-act-pair · gpt-3.5-turbo

    references to confessions or acts of admitting wrongdoing

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Negative Logits
     impacted
    -0.81
     Phase
    -0.81
     impacts
    -0.80
     Impact
    -0.79
     impact
    -0.79
    impact
    -0.68
     Riders
    -0.67
    raviolet
    -0.67
    agnetic
    -0.66
    alf
    -0.64
    POSITIVE LOGITS
     confess
    3.28
     confessed
    3.14
     confession
    3.05
     confessions
    2.80
     repent
    1.38
     profess
    1.30
     repentance
    1.23
     incrim
    1.15
     contr
    1.15
     admitting
    1.13
    Activations Density 0.031%

    No Known Activations