© 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 EvalsExportsinterp-engineNEWAPI 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. 11-RES-JB
    11. 13225
    Prev
    Next
    INDEX
    Explanations

    phrases related to media coverage such as "covered," "reporting," "exaggerate," and "bias."

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to media coverage and reporting practices

    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
     teasp
    -0.82
    à¼
    -0.74
    Timer
    -0.72
    starter
    -0.71
    retty
    -0.70
    externalActionCode
    -0.68
    timer
    -0.66
     guiActiveUnfocused
    -0.65
     bookmark
    -0.64
    Iterator
    -0.64
    POSITIVE LOGITS
     scandals
    1.28
     controversies
    1.17
     tragedies
    1.10
     atrocities
    1.09
     injust
    1.08
     abuses
    1.07
     allegations
    1.07
     wrongdoing
    1.01
     incidents
    1.01
     events
    0.98
    Activations Density 0.413%

    No Known Activations