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. 11-RES-JB
    6. 19181
    Prev
    Next
    INDEX
    Explanations

    phrases discussing the physical components and condition of a vehicle

    oai_token-act-pair · gpt-3.5-turbo

    technical descriptions related to components and systems

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Merit
    -0.92
    eday
    -0.85
    nesday
    -0.77
    Pages
    -0.73
    elist
    -0.72
    rencies
    -0.72
    leground
    -0.71
    otle
    -0.71
    conom
    -0.71
    lees
    -0.70
    POSITIVE LOGITS
     propulsion
    1.53
     cockpit
    1.49
     propell
    1.39
     parachute
    1.35
     parach
    1.31
     prope
    1.30
     engines
    1.29
     fu
    1.28
     onboard
    1.25
     aft
    1.24
    Activations Density 0.223%

    No Known Activations