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

    Neuronpedia

    Jacobian LensNEW
    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. 8-RES-JB
    11. 10316
    Prev
    Next
    INDEX
    Explanations

    phrases related to specific items or concepts, often followed by descriptive words or adjectives

    oai_token-act-pair · gpt-3.5-turbo

    references to "the" in various contexts indicating focus or emphasis

    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
    âĢİ
    -0.73
    eus
    -0.62
    IPS
    -0.61
     Bridges
    -0.61
    ATURES
    -0.61
    """
    -0.60
    riages
    -0.60
    owicz
    -0.59
    ledge
    -0.59
     Rutherford
    -0.58
    POSITIVE LOGITS
     utmost
    0.94
     aforementioned
    0.90
     latter
    0.85
     dreaded
    0.84
     same
    0.84
    same
    0.81
     finest
    0.78
     infamous
    0.77
     usual
    0.71
     elusive
    0.70
    Activations Density 0.347%

    No Known Activations