© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 8-RES-JB
    11. 4562
    Prev
    Next
    INDEX
    Explanations

    phrases related to celestial bodies, particularly Saturn and its moons

    oai_token-act-pair · gpt-3.5-turbo

    references to celestial bodies and mythological figures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    lier
    -0.83
    riter
    -0.83
    ocard
    -0.77
    stand
    -0.75
    erate
    -0.74
    think
    -0.74
    galitarian
    -0.72
    erm
    -0.70
    rar
    -0.70
    manac
    -0.68
    POSITIVE LOGITS
     Mercury
    0.86
     Saturn
    0.84
     Asc
    0.81
     Uran
    0.80
    arium
    0.77
     Neptune
    0.76
    images
    0.76
     Pluto
    0.75
    alia
    0.74
     spacecraft
    0.74
    Activations Density 0.097%

    No Known Activations