© 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. 0-RES-JB
    11. 10361
    Prev
    Next
    INDEX
    Explanations

    phrases related to creating or causing something

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "make."

    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
    mania
    -0.70
    rist
    -0.68
    heit
    -0.65
    ilion
    -0.65
    hered
    -0.65
    meier
    -0.63
    cest
    -0.62
    inar
    -0.60
     Ended
    -0.59
    edient
    -0.57
    POSITIVE LOGITS
     make
    2.70
    make
    2.24
     MAKE
    2.07
     Make
    1.96
    Make
    1.87
     create
    1.53
     makes
    1.45
     give
    1.44
     ensure
    1.41
     prepare
    1.39
    Activations Density 0.069%

    No Known Activations