© 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. 4-RES-JB
    11. 14389
    Prev
    Next
    INDEX
    Explanations

    abstract ideas or concepts

    oai_token-act-pair · gpt-3.5-turbo

    key terms related to abstract concepts and significant actions

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

    No Comments

    Negative Logits
    _>
    -0.67
     Doors
    -0.61
    ummies
    -0.61
    racuse
    -0.59
    photos
    -0.57
    inas
    -0.56
    ergy
    -0.55
    Ingredients
    -0.54
    Assembly
    -0.54
    party
    -0.54
    POSITIVE LOGITS
     extends
    0.78
     represents
    0.71
     belongs
    0.70
     assumes
    0.70
     lacks
    0.68
     encompasses
    0.68
     may
    0.67
     arose
    0.66
     lasted
    0.65
     coincides
    0.65
    Activations Density 0.288%

    No Known Activations