© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 8-RES-JB
    11. 8237
    Prev
    Next
    INDEX
    Explanations

    phrases related to sharing personal opinions and assurances, often including phrases like "believe me," "trust me," and "I promise."

    oai_token-act-pair · gpt-3.5-turbo

    affirmations and expressions of trust or promises

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    fund
    -0.64
     disemb
    -0.61
     basket
    -0.60
    etting
    -0.60
     vanish
    -0.60
    Applications
    -0.58
     disappear
    -0.57
    Fund
    -0.57
    letal
    -0.56
     deposition
    -0.56
    POSITIVE LOGITS
    eno
    0.74
    ELY
    0.63
    oso
    0.60
    heimer
    0.60
    agers
    0.58
    alan
    0.56
    irin
    0.56
     admit
    0.56
    thouse
    0.55
    ggie
    0.55
    Activations Density 0.112%

    No Known Activations