© 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. 12-RES-JB
    11. 4874
    Prev
    Next
    INDEX
    Explanations

    mentions of sports incidents and actions involving penalties and suspensions

    oai_token-act-pair · gpt-3.5-turbo
    Triggered by @bot

    instances of disciplinary actions or notable decisions in sports contexts

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 12-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_post
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.11.hook_resid_post
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
    eneg
    -0.93
    arse
    -0.82
    license
    -0.82
    quished
    -0.80
    asters
    -0.80
     nib
    -0.79
    Saharan
    -0.77
    parse
    -0.75
    Magikarp
    -0.74
     sorce
    -0.73
    POSITIVE LOGITS
     Hill
    2.14
    Hill
    2.07
     Hall
    1.89
    Hall
    1.79
     Wall
    1.62
     Halls
    1.54
    Wall
    1.51
    hall
    1.44
     Walls
    1.40
     Holl
    1.32
    Activations Density 0.277%

    No Known Activations