Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 7-RES-JB
    6. 7530
    Prev
    Next
    INDEX
    Explanations

    references to specific events or controversial topics such as disasters, political issues, and social conflicts

    oai_token-act-pair · gpt-3.5-turbo

    connections involving the word "and."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 7-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    7
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    lit
    -0.74
    pi
    -0.67
    iless
    -0.65
    unci
    -0.62
    mus
    -0.62
    look
    -0.62
    XY
    -0.62
    obe
    -0.61
    live
    -0.61
    lamm
    -0.61
    POSITIVE LOGITS
     thereby
    1.25
     consequently
    1.12
     thus
    1.08
     hence
    1.03
     therefore
    0.90
     consequ
    0.90
     thence
    0.89
     subsequently
    0.88
     furthermore
    0.87
     thereafter
    0.87
    Activations Density 0.438%

    No Known Activations