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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 11824
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "Bear" with varying intensifies at 9 and 10

    oai_token-act-pair · gpt-3.5-turbo

    references to "bear" 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
     surg
    -0.74
    ruct
    -0.65
     pseudo
    -0.65
    ractive
    -0.65
     prost
    -0.65
     xen
    -0.64
    lde
    -0.64
     Lumpur
    -0.64
     electr
    -0.63
    idem
    -0.63
    POSITIVE LOGITS
     Bear
    3.98
    Bear
    3.00
    bear
    2.35
     bear
    2.21
     Bears
    2.10
     bears
    1.78
     Grizz
    1.46
     Panda
    1.46
     Bearing
    1.36
     Tiger
    1.25
    Activations Density 0.013%

    No Known Activations