© 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. 8-RES-JB
    11. 19250
    Prev
    Next
    INDEX
    Explanations

    terms related to managing, dealing with, or taking care of tasks or situations

    oai_token-act-pair · gpt-3.5-turbo

    variations of the word "handle" and its forms

    oai_token-act-pair · gpt-4o-miniTriggered 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
    Not in Any Lists

    No Comments

    Negative Logits
    hemat
    -0.74
    ischer
    -0.73
    esis
    -0.72
    mpeg
    -0.71
    iary
    -0.64
    urt
    -0.64
    inet
    -0.63
    zyme
    -0.62
    zai
    -0.61
    aez
    -0.61
    POSITIVE LOGITS
    bars
    1.15
     handles
    0.71
     handle
    0.70
     handled
    0.66
     sensitive
    0.66
     luggage
    0.65
     emergencies
    0.64
     paperwork
    0.64
     mishand
    0.64
     responsibilities
    0.61
    Activations Density 0.041%

    No Known Activations