© 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. 0-RES-JB
    11. 22279
    Prev
    Next
    INDEX
    Explanations

    proper nouns or terms related to conflicts, wars, or battles

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "fought" 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
     sample
    -0.66
     excess
    -0.64
     plug
    -0.63
     outlet
    -0.62
     applications
    -0.62
     slip
    -0.62
     perfection
    -0.62
     maximum
    -0.61
     Express
    -0.61
     error
    -0.61
    POSITIVE LOGITS
     fought
    3.61
     battled
    2.13
     waged
    2.00
     fights
    1.69
     fight
    1.66
     resisted
    1.58
    Fight
    1.58
    fight
    1.50
     defended
    1.50
     competed
    1.46
    Activations Density 0.014%

    No Known Activations