© 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. 3-RES-JB
    11. 5601
    Prev
    Next
    INDEX
    Explanations

    the word "girl" with various levels of emphasis

    oai_token-act-pair · gpt-3.5-turbo

    references to characters and elements associated with "girl" as well as related names

    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
    ħĭ
    -0.71
    QUIRE
    -0.60
     imposing
    -0.59
     demanding
    -0.59
     courageous
    -0.58
     smart
    -0.58
     punishing
    -0.58
     pledging
    -0.57
     marrow
    -0.57
     presumably
    -0.57
    POSITIVE LOGITS
    oin
    0.95
    onge
    0.93
    iffe
    0.92
    itudinal
    0.92
    ibrary
    0.89
    oop
    0.89
    atches
    0.86
    ament
    0.85
    oom
    0.84
    itude
    0.84
    Activations Density 0.010%

    No Known Activations