© 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. 10-RES-JB
    11. 4384
    Prev
    Next
    INDEX
    Explanations

    superlatives describing something as the best or the worst that has ever occurred

    oai_token-act-pair · gpt-3.5-turbo

    superlative expressions or phrases indicating something is the greatest or most significant

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ento
    -0.76
    avascript
    -0.71
    oren
    -0.70
    ose
    -0.65
    asionally
    -0.65
    ERAL
    -0.65
    icio
    -0.63
    ãĥ´ãĤ¡
    -0.62
    ofer
    -0.61
    Reference
    -0.61
    POSITIVE LOGITS
     conceived
    1.21
     recorded
    1.18
     devised
    1.14
     assembled
    1.05
     created
    1.05
     produced
    1.02
     seen
    1.02
     achieved
    0.97
     bestowed
    0.97
     invented
    0.97
    Activations Density 0.033%

    No Known Activations