© 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. 3433
    Prev
    Next
    INDEX
    Explanations

    the word "Orion" and related terms such as "constellation" or specific cultural references to Orion

    oai_token-act-pair · gpt-3.5-turbo

    references to celestial bodies and specific entities related to space and astronomy

    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
    aign
    -0.85
    lished
    -0.84
    eteria
    -0.83
    lasses
    -0.80
    ern
    -0.75
    sheets
    -0.75
    ablishment
    -0.74
    inity
    -0.73
    galitarian
    -0.73
    arten
    -0.71
    POSITIVE LOGITS
     constellation
    1.03
     Orion
    0.94
     Nebula
    0.90
     Predator
    0.83
     Odyssey
    0.76
     tremend
    0.75
     Centauri
    0.74
     Polaris
    0.72
     coil
    0.71
    atives
    0.70
    Activations Density 0.029%

    No Known Activations