© 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. 8-RES-JB
    11. 869
    Prev
    Next
    INDEX
    Explanations

    phrases related to a specific name or location, specifically "Budd" or "Sudd" the latter of which appears more frequently

    oai_token-act-pair · gpt-3.5-turbo

    the word "udd" in various forms and contexts throughout the document

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    lda
    -0.74
    pora
    -0.66
     understatement
    -0.63
     lobe
    -0.61
    izons
    -0.59
    SPONSORED
    -0.58
     Freedom
    -0.57
    ======
    -0.57
     Dome
    -0.55
     lapse
    -0.55
    POSITIVE LOGITS
    ington
    1.00
    ernaut
    0.98
    issance
    0.98
    ying
    0.94
    ership
    0.93
    erers
    0.91
    orld
    0.90
    itional
    0.90
    ski
    0.88
    ruck
    0.85
    Activations Density 0.024%

    No Known Activations