© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 8-RES-JB
    6. 9858
    Prev
    Next
    INDEX
    Explanations

    emails or messages with a specific format such as quotations and arrows

    oai_token-act-pair · gpt-3.5-turbo

    specific formatting and symbols indicative of structured data or dialogue in a communication context

    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
    •contextual features
    Fires on buffer text, especially ">> >>" or "> >"

    No Comments

    Negative Logits
    ÄŁ
    -0.85
    issance
    -0.79
    aina
    -0.78
    acia
    -0.77
    haw
    -0.76
    oat
    -0.75
    onia
    -0.75
    osc
    -0.73
    adesh
    -0.73
    atre
    -0.73
    POSITIVE LOGITS
     ********************************
    0.99
    ----------------------------------------------------------------
    0.95
     =================================================================
    0.87
    ================================
    0.79
    ================================================================
    0.76
    ********************************
    0.75
     Copyright
    0.71
     CONT
    0.69
    ////////////////////////////////
    0.69
     NOTICE
    0.68
    Activations Density 0.086%

    No Known Activations