© 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. 10-RES-JB
    11. 7919
    Prev
    Next
    INDEX
    Explanations

    phrases related to expressing sympathy or concern for individuals in difficult situations

    oai_token-act-pair · gpt-3.5-turbo

    references to victims and their families in contexts of loss or tragedy

    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
    Sty
    -0.68
    quickShipAvailable
    -0.66
     Heist
    -0.66
     Style
    -0.65
    ascar
    -0.64
    Style
    -0.63
    Official
    -0.63
    Sales
    -0.63
    ħĭ
    -0.62
     Onion
    -0.60
    POSITIVE LOGITS
     impacted
    1.31
     suffering
    1.22
     victimized
    1.19
     affected
    1.18
     stricken
    1.15
     trapped
    1.12
     stranded
    1.12
     devastated
    1.10
     afflicted
    1.09
     struggling
    1.08
    Activations Density 0.338%

    No Known Activations