© 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. 1-RES-JB
    11. 12941
    Prev
    Next
    INDEX
    Explanations

    phrases indicating future intentions or predictions

    oai_token-act-pair · gpt-3.5-turbo

    the phrase "gonna" which indicates future intentions or predictions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    hips
    -0.84
    icipated
    -0.80
    operated
    -0.78
    lying
    -0.77
    unc
    -0.76
    aments
    -0.76
    mere
    -0.75
    ellation
    -0.73
    lessly
    -0.72
    template
    -0.71
    POSITIVE LOGITS
     gonna
    1.10
     wanna
    0.97
     tha
    0.87
     gotta
    0.86
     nab
    0.80
     snag
    0.79
     buckle
    0.78
     smack
    0.77
     msec
    0.76
     notch
    0.76
    Activations Density 0.004%

    No Known Activations