© 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. 1-RES-JB
    11. 13472
    Prev
    Next
    INDEX
    Explanations

    words related to being underwater and submerged

    oai_token-act-pair · gpt-3.5-turbo

    terms related to underwater environments and aquatic themes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ========
    -0.93
    aign
    -0.89
    ============
    -0.86
    xx
    -0.86
     CLSID
    -0.83
    Brand
    -0.79
    ====
    -0.79
    ================
    -0.78
    present
    -0.78
    xxxxxxxx
    -0.77
    POSITIVE LOGITS
     underwater
    1.44
     mathemat
    1.10
     diving
    1.04
     submerged
    0.98
    ¥ŀ
    0.94
     aquarium
    0.94
    conservancy
    0.94
     landsl
    0.92
    eatures
    0.91
     eleph
    0.90
    Activations Density 0.005%

    No Known Activations