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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 0-RES-JB
    6. 9215
    Prev
    Next
    INDEX
    Explanations

    phrases where the letters "hu" are present and possibly associated with names or specific terms

    oai_token-act-pair · gpt-3.5-turbo

    the presence of names or terms related to specific individuals or characters

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Anchorage
    -0.71
     Camden
    -0.70
     stat
    -0.65
     BASE
    -0.64
     Brooke
    -0.64
     control
    -0.63
     keys
    -0.62
     parenting
    -0.62
     Progress
    -0.61
     Keys
    -0.61
    POSITIVE LOGITS
    hu
    4.59
    hus
    1.93
    hai
    1.70
    hua
    1.68
    hou
    1.57
    Hu
    1.52
    ho
    1.48
    hi
    1.47
    hur
    1.46
    ha
    1.46
    Activations Density 0.009%

    No Known Activations