© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 4-RES-JB
    11. 10587
    Prev
    Next
    INDEX
    Explanations

    phrases related to enabling or making something possible

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "enable" and its variations, indicating capability or potential

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

    No Comments

    Negative Logits
    boy
    -0.80
     Rite
    -0.74
     Nose
    -0.67
     Warcraft
    -0.67
    den
    -0.65
    bane
    -0.64
     Soda
    -0.62
    worn
    -0.62
    mal
    -0.61
     GEAR
    -0.61
    POSITIVE LOGITS
    ments
    0.84
    Reviewer
    0.83
    auga
    0.81
    iences
    0.80
    ment
    0.80
    untarily
    0.79
    cially
    0.77
    initions
    0.73
    htaking
    0.73
    ances
    0.72
    Activations Density 0.027%

    No Known Activations