© 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. 0-RES-JB
    11. 12496
    Prev
    Next
    INDEX
    Explanations

    phrases related to actions or activities, particularly those that involve taking or holding something

    oai_token-act-pair · gpt-3.5-turbo

    forms of the word "taking"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     aggregate
    -0.68
    ator
    -0.67
     ad
    -0.66
     warn
    -0.66
     hype
    -0.64
     feed
    -0.64
     mang
    -0.63
     upgraded
    -0.63
     adam
    -0.63
     bat
    -0.63
    POSITIVE LOGITS
    taking
    3.90
    bringing
    1.38
    holding
    1.30
    giving
    1.25
    making
    1.22
    Taking
    1.21
    raising
    1.19
    telling
    1.15
    calling
    1.09
    getting
    1.04
    Activations Density 0.011%

    No Known Activations