© 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. 10-RES-JB
    11. 3334
    Prev
    Next
    INDEX
    Explanations

    quotes or reported speech, especially from interviews or court cases

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "told."

    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
     ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³
    -0.97
    mite
    -0.78
    ample
    -0.77
    pmwiki
    -0.75
    prime
    -0.73
    aband
    -0.73
    é¾įåĸļ士
    -0.73
    ãĤ¼ãĤ¦ãĤ¹
    -0.73
    adesh
    -0.72
    Interstitial
    -0.70
    POSITIVE LOGITS
     BuzzFeed
    1.30
     reporters
    1.25
     HuffPost
    1.15
     CNN
    1.06
     TMZ
    1.05
     investigators
    1.02
     me
    1.01
     NBC
    0.99
     Reuters
    0.99
     CBS
    0.98
    Activations Density 0.062%

    No Known Activations