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. 23656
    Prev
    Next
    INDEX
    Explanations

    instances of the word "mark" used in a context related to accolades or milestones

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "mark" and its variations, indicating a focus on important milestones or indicators

    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
    itely
    -0.65
    esta
    -0.63
    Pros
    -0.62
     Investigations
    -0.62
    aghetti
    -0.62
    sle
    -0.62
    olia
    -0.60
     Pros
    -0.60
    alli
    -0.60
    qua
    -0.60
    POSITIVE LOGITS
     mark
    4.16
     marks
    2.57
    mark
    2.32
    marks
    2.10
     marking
    1.83
    marked
    1.64
     Marks
    1.55
     MARK
    1.53
    Mark
    1.52
     Mark
    1.48
    Activations Density 0.015%

    No Known Activations