© 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. 0-RES-JB
    11. 8121
    Prev
    Next
    INDEX
    Explanations

    proper nouns related to rocketry or space exploration

    oai_token-act-pair · gpt-3.5-turbo

    references to a specific car model named "Rocket."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     slavery
    -0.85
     trough
    -0.79
     seiz
    -0.79
     regress
    -0.78
     compr
    -0.78
     vic
    -0.77
     pleas
    -0.77
    SPONSORED
    -0.74
     subt
    -0.73
    ãĥĩãĤ£
    -0.71
    POSITIVE LOGITS
     Rocket
    3.79
    Rocket
    2.99
     Rockets
    1.87
    rocket
    1.71
     Atomic
    1.41
     Missile
    1.34
     Psy
    1.33
     Squirrel
    1.26
     Robot
    1.24
     ROCK
    1.23
    Activations Density 0.038%

    No Known Activations