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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    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. 10-RES-JB
    6. 3454
    Prev
    Next
    INDEX
    Explanations

    words ending with a specific symbol ("Ċ") and numbers adjacent to these symbols

    oai_token-act-pair · gpt-3.5-turbo

    references to critical issues and concerns

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Sob
    -0.82
     Rebel
    -0.81
     Conf
    -0.80
     Fest
    -0.79
     Olympic
    -0.79
     Font
    -0.78
     Quart
    -0.78
     Dem
    -0.77
     Tri
    -0.77
     Reaper
    -0.76
    POSITIVE LOGITS
    Secondly
    1.32
    Anonymous
    1.31
    why
    1.29
    another
    1.28
    Advertisements
    1.27
    these
    1.27
    then
    1.27
    THE
    1.26
    CONCLUS
    1.25
    Reviewer
    1.24
    Activations Density 0.448%

    No Known Activations