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. 10179
    Prev
    Next
    INDEX
    Explanations

    verbs or adjectives that emphasize or highlight something

    oai_token-act-pair · gpt-3.5-turbo

    words related to emphasizing or underscoring important points

    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
     combust
    -0.76
     sabot
    -0.64
     contrace
    -0.64
     sanctuary
    -0.63
     fallout
    -0.62
     tampering
    -0.62
     naughty
    -0.61
     unprotected
    -0.61
     sidel
    -0.60
     hiber
    -0.60
    POSITIVE LOGITS
    idate
    1.01
    idated
    1.00
    ribes
    0.97
    itating
    0.96
    orical
    0.95
    ications
    0.95
    idates
    0.95
    icating
    0.92
    itated
    0.92
    ising
    0.90
    Activations Density 0.076%

    No Known Activations