© 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. 3-RES-JB
    11. 690
    Prev
    Next
    INDEX
    Explanations

    references to physical structures or objects, particularly related to size and installation

    oai_token-act-pair · gpt-3.5-turbo

    references to "frames" or "frame" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    tical
    -0.78
    zai
    -0.77
    iciency
    -0.77
    withstanding
    -0.76
    icably
    -0.76
    inse
    -0.75
    emale
    -0.74
    itable
    -0.74
    utenberg
    -0.74
    inen
    -0.73
    POSITIVE LOGITS
     frames
    1.03
    buffer
    1.02
     Frames
    0.93
     frame
    0.92
    border
    0.83
     Frame
    0.83
    builder
    0.79
    frames
    0.75
    Frames
    0.73
     Breaker
    0.73
    Activations Density 0.007%

    No Known Activations