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

    Neuronpedia

    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. 7-RES-JB
    11. 18029
    Prev
    Next
    INDEX
    Explanations

    specific targets or entities that are the focus of some action or intention

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "targeted."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 7-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    7
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    fo
    -0.77
    batch
    -0.72
    rar
    -0.72
    loo
    -0.71
    SourceFile
    -0.70
    ston
    -0.66
    Else
    -0.66
    aug
    -0.64
    shire
    -0.64
    iHUD
    -0.64
    POSITIVE LOGITS
     targeting
    0.87
     squarely
    0.84
     targeted
    0.82
     targets
    0.81
    ted
    0.79
     killings
    0.75
    ãĤ°
    0.70
     towards
    0.70
     ads
    0.69
     target
    0.69
    Activations Density 0.038%

    No Known Activations