© 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. 11-RES-JB
    11. 20597
    Prev
    Next
    INDEX
    Explanations

    instances where the word "keen" is used in a positive context or to express enthusiasm

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "keen" used to express eagerness or desire

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ioxide
    -0.62
    essage
    -0.60
    channelAvailability
    -0.58
    miah
    -0.57
    cano
    -0.57
    ructure
    -0.56
    _.
    -0.55
    ublic
    -0.55
    天
    -0.55
    ailable
    -0.55
    POSITIVE LOGITS
    ly
    1.21
    ness
    1.04
     to
    0.86
    footed
    0.78
    nesses
    0.78
     enough
    0.77
    est
    0.76
     listeners
    0.74
     buyers
    0.70
    ingly
    0.70
    Activations Density 0.061%

    No Known Activations