Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 17577
    Prev
    Next
    INDEX
    Explanations

    phrases related to being caught or captured

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "caught" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    determination
    -0.67
    larg
    -0.66
    ointment
    -0.65
     Pledge
    -0.64
     MOD
    -0.64
    atives
    -0.64
    ient
    -0.64
     Independence
    -0.63
     shrink
    -0.62
    itability
    -0.62
    POSITIVE LOGITS
     caught
    3.60
     catches
    1.93
     captured
    1.74
     catch
    1.70
     catching
    1.66
     intercepted
    1.45
     hauled
    1.38
     busted
    1.37
    aught
    1.32
     trapped
    1.32
    Activations Density 0.018%

    No Known Activations