© 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. 11598
    Prev
    Next
    INDEX
    Explanations

    mentions of the specific word "rag" with varying contexts or emphasis

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "rag."

    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
     Channel
    -0.64
     rating
    -0.64
     condos
    -0.63
     Jerome
    -0.63
     broadcast
    -0.62
     Livingston
    -0.62
     markets
    -0.60
     Cook
    -0.60
     claims
    -0.59
     Newman
    -0.59
    POSITIVE LOGITS
    rag
    4.87
     rag
    1.46
    RAG
    1.36
     Rag
    1.28
    ragon
    1.22
    rab
    1.17
    lag
    1.12
    rane
    1.09
    rig
    1.08
    rage
    1.07
    Activations Density 0.011%

    No Known Activations