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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 1-RES-JB
    11. 20039
    Prev
    Next
    INDEX
    Explanations

    verbs related to catching someone in the act

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "caught."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     pronoun
    -0.81
     pronouns
    -0.77
    guided
    -0.71
    xual
    -0.69
     reconc
    -0.68
    arth
    -0.67
    enture
    -0.67
    ensible
    -0.67
    umably
    -0.65
     helicop
    -0.64
    POSITIVE LOGITS
     caught
    0.94
    ãĤµ
    0.87
    ãĥĩãĤ£
    0.79
     Luffy
    0.75
    Frames
    0.71
     catches
    0.70
     Rai
    0.70
    ãĤ·ãĥ£
    0.69
     Tracks
    0.69
    ãĤĮ
    0.69
    Activations Density 0.010%

    No Known Activations