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. 9356
    Prev
    Next
    INDEX
    Explanations

    words related to people's names, especially the name "Mark"

    oai_token-act-pair · gpt-3.5-turbo

    mentions of the name "Mark."

    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
    opian
    -0.93
    qua
    -0.76
     corrid
    -0.75
    querque
    -0.73
    allo
    -0.68
     flourishing
    -0.65
    ifferent
    -0.65
    swer
    -0.64
    cu
    -0.64
    ushi
    -0.64
    POSITIVE LOGITS
     Mark
    3.44
    Mark
    2.71
    mark
    2.00
     MARK
    1.93
     mark
    1.85
     Marks
    1.74
    marks
    1.59
     Matthew
    1.49
     Luke
    1.45
     Marc
    1.39
    Activations Density 0.016%

    No Known Activations