© 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. 992
    Prev
    Next
    INDEX
    Explanations

    words related to high expertise or qualification

    oai_token-act-pair · gpt-3.5-turbo

    the adjective "highly" in various contexts

    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
     Nightmares
    -0.73
    psons
    -0.72
     Rite
    -0.68
     Chaser
    -0.67
     Ashe
    -0.67
    ppo
    -0.67
     Dare
    -0.66
     halves
    -0.66
     Parenthood
    -0.66
    ories
    -0.66
    POSITIVE LOGITS
     regarded
    1.07
     educated
    0.95
     rated
    0.95
     unlikely
    0.94
     prized
    0.91
     anticipated
    0.91
     publicized
    0.89
     improbable
    0.87
     indebted
    0.83
     esteemed
    0.83
    Activations Density 0.025%

    No Known Activations