© 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. 2-RES-JB
    11. 3198
    Prev
    Next
    INDEX
    Explanations

    instances of the word "juice", with a particularly high focus on occurrences of "juice" followed by a numerical value

    oai_token-act-pair · gpt-3.5-turbo

    references to various types of juice

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    gotten
    -0.71
    sighted
    -0.71
    pict
    -0.70
    rict
    -0.69
    etheless
    -0.69
    military
    -0.68
    certain
    -0.67
    severe
    -0.65
    humane
    -0.63
    ear
    -0.62
    POSITIVE LOGITS
     Juice
    1.26
     juice
    1.25
     juices
    1.12
    xtap
    0.89
    wine
    0.88
    weed
    0.85
     Beans
    0.85
    ée
    0.82
     conco
    0.81
    fruit
    0.80
    Activations Density 0.007%

    No Known Activations