© 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. 4-RES-JB
    11. 14705
    Prev
    Next
    INDEX
    Explanations

    words related to engines

    oai_token-act-pair · gpt-3.5-turbo

    mentions of engines and their characteristics

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

    No Comments

    Negative Logits
    laus
    -0.97
    imedia
    -0.84
    istani
    -0.83
    soever
    -0.82
    Ń·
    -0.77
    atana
    -0.73
     Childhood
    -0.73
    oslov
    -0.71
    lyak
    -0.71
    dos
    -0.71
    POSITIVE LOGITS
     powering
    0.93
     compartment
    0.89
     braking
    0.85
     engine
    0.84
     ignition
    0.81
     horsepower
    0.78
     engines
    0.78
     powered
    0.78
     bay
    0.77
     roared
    0.76
    Activations Density 0.013%

    No Known Activations