© 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. 10230
    Prev
    Next
    INDEX
    Explanations

    the word "though" as a conjunction or qualifier in sentences

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "though" used to express contrast or qualification

    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
    asus
    -0.67
    enter
    -0.61
    izational
    -0.61
    ivery
    -0.60
    inated
    -0.59
    lees
    -0.59
    izes
    -0.58
    pta
    -0.58
    brate
    -0.58
    arted
    -0.58
    POSITIVE LOGITS
    tons
    0.96
     admittedly
    0.87
    ts
    0.84
    ullah
    0.76
     beware
    0.76
     thankfully
    0.74
     interestingly
    0.74
     fortunately
    0.72
    abouts
    0.70
    still
    0.70
    Activations Density 0.043%

    No Known Activations