© 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. 2772
    Prev
    Next
    INDEX
    Explanations

    superlatives related to quantities or levels

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "lowest."

    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
    arry
    -0.64
    nox
    -0.64
    aus
    -0.60
    uther
    -0.59
    enko
    -0.59
    kamp
    -0.59
    craft
    -0.58
    zes
    -0.56
     Invasion
    -0.56
    ifa
    -0.56
    POSITIVE LOGITS
     lowest
    3.50
     highest
    2.42
    highest
    2.09
     smallest
    1.97
     cheapest
    1.93
     weakest
    1.85
     widest
    1.85
     shortest
    1.78
     Highest
    1.71
     heaviest
    1.69
    Activations Density 0.008%

    No Known Activations