© 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. 4-RES-JB
    11. 14810
    Prev
    Next
    INDEX
    Explanations

    mentions of the word "Tap" with various capitalizations

    oai_token-act-pair · gpt-3.5-turbo

    the term "tap" in various contexts, indicating its focus on topics related to tapping, whether for beverages or other uses

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

    No Comments

    Negative Logits
    theless
    -0.79
    ften
    -0.76
     CVE
    -0.74
     AMERICA
    -0.73
    Reloaded
    -0.72
    etheless
    -0.70
    Downloadha
    -0.68
    ãĤ¼ãĤ¦ãĤ¹
    -0.67
    IGHTS
    -0.67
    sbm
    -0.65
    POSITIVE LOGITS
    Tap
    0.94
    acity
    0.92
     tap
    0.90
    tap
    0.88
    rop
    0.87
    ering
    0.82
    deck
    0.81
    itol
    0.81
    ro
    0.81
    ropri
    0.81
    Activations Density 0.007%

    No Known Activations