© 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. 13257
    Prev
    Next
    INDEX
    Explanations

    phrases where something or someone played a crucial role in a positive outcome or achievement

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "instrumental" and its variations

    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
     Refresh
    -0.72
    Sense
    -0.67
    kai
    -0.66
     refresh
    -0.65
    rist
    -0.63
     Scroll
    -0.61
    Clear
    -0.61
    wall
    -0.60
    rx
    -0.59
     Conan
    -0.59
    POSITIVE LOGITS
     instrumental
    3.76
    strument
    1.74
     Instrument
    1.37
     instrument
    1.30
     instruments
    1.25
     vocal
    1.23
     orche
    1.21
     integral
    1.17
     pivotal
    1.04
     orchestr
    1.01
    Activations Density 0.021%

    No Known Activations