© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 10-RES-JB
    11. 1244
    Prev
    Next
    INDEX
    Explanations

    phrases introducing a change or transition in a text

    oai_token-act-pair · gpt-3.5-turbo

    phrases that start with "Over" followed by a numerical quantity and a period of time

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Haram
    -0.61
    ittal
    -0.58
     guilty
    -0.56
     Express
    -0.56
     notation
    -0.56
    BLE
    -0.55
     DI
    -0.55
     belong
    -0.55
    itaire
    -0.55
    MSN
    -0.54
    POSITIVE LOGITS
    whelming
    1.41
    whel
    1.30
    lap
    1.24
    haul
    1.23
    arching
    1.21
    seeing
    1.19
    coming
    1.17
    dose
    1.16
    come
    1.12
    heard
    1.12
    Activations Density 0.038%

    No Known Activations