© 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. 8-RES-JB
    11. 13135
    Prev
    Next
    INDEX
    Explanations

    mentions of specific academic years in relation to a person's first year in college or high school

    oai_token-act-pair · gpt-3.5-turbo

    references to academic years, particularly freshman and sophomore classifications

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
    ichick
    -0.86
    rosis
    -0.80
    abulary
    -0.79
    pots
    -0.76
     annexed
    -0.76
    igree
    -0.70
    tering
    -0.70
    orius
    -0.70
    leased
    -0.69
    eful
    -0.69
    POSITIVE LOGITS
     dorm
    1.01
    uates
    0.98
     slump
    0.93
     freshman
    0.88
    omore
    0.86
     freshmen
    0.86
     soph
    0.86
    arsity
    0.83
     sophomore
    0.82
     semester
    0.81
    Activations Density 0.040%

    No Known Activations