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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 7-RES-JB
    6. 12865
    Prev
    Next
    INDEX
    Explanations

    phrases or words indicating an increase or intensification in some kind of activity, effort, or action

    oai_token-act-pair · gpt-3.5-turbo

    instances of the phrase "step up."

    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
    iol
    -0.73
    elf
    -0.72
    uel
    -0.64
    eport
    -0.62
    \\\\\\\\
    -0.62
     Declaration
    -0.62
    Ñı
    -0.60
    iled
    -0.60
     Crate
    -0.60
    esm
    -0.59
    POSITIVE LOGITS
    river
    0.83
    stairs
    0.80
    grading
    0.73
    pressure
    0.69
     efforts
    0.68
    olicy
    0.67
    assault
    0.67
     srfAttach
    0.66
    grades
    0.65
    rates
    0.65
    Activations Density 0.030%

    No Known Activations