Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 9-RES-JB
    6. 12341
    Prev
    Next
    INDEX
    Explanations

    geometric shapes and properties

    oai_token-act-pair · gpt-3.5-turbo

    concepts related to celestial mechanics and orbital dynamics

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    9
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    pkg
    -0.98
    ufact
    -0.88
     Pharma
    -0.86
    nsics
    -0.81
     Abuse
    -0.76
     Tobacco
    -0.72
     urgently
    -0.70
    priv
    -0.70
     Priv
    -0.70
     Drinking
    -0.68
    POSITIVE LOGITS
     perpendicular
    1.48
     symmetry
    1.42
     axis
    1.40
     curves
    1.40
     orbits
    1.32
     curved
    1.31
     radial
    1.29
     horizontal
    1.29
     gradient
    1.29
     symm
    1.28
    Activations Density 0.655%

    No Known Activations