© 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. 3-RES-JB
    11. 16186
    Prev
    Next
    INDEX
    Explanations

    phrases related to upgrades or additional features offered at a higher level

    oai_token-act-pair · gpt-3.5-turbo

    the word "Plus" in various contexts, indicating additional features or benefits

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    adr
    -0.99
    taboola
    -0.84
    stem
    -0.81
    actory
    -0.80
    itory
    -0.79
    phies
    -0.78
    amental
    -0.78
    perature
    -0.78
    fell
    -0.77
    vironment
    -0.76
    POSITIVE LOGITS
    Plus
    1.11
     Plus
    1.01
     PLUS
    0.93
     Bonus
    0.74
     Nico
    0.73
     Plenty
    0.73
    ++++
    0.72
     Ukrain
    0.68
    +++
    0.67
    minus
    0.67
    Activations Density 0.008%

    No Known Activations