© 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. 6777
    Prev
    Next
    INDEX
    Explanations

    words related to softness or gentleness

    oai_token-act-pair · gpt-3.5-turbo

    the word "soft" in various contexts

    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
     Telegraph
    -0.81
    ITNESS
    -0.76
     Haunted
    -0.72
    andals
    -0.71
    USS
    -0.69
    drivers
    -0.68
     Bam
    -0.68
     Amtrak
    -0.68
     Metropolitan
    -0.67
    ivals
    -0.67
    POSITIVE LOGITS
     palate
    1.17
    soft
    1.06
    ener
    0.99
    ening
    0.99
     soft
    0.95
     softer
    0.91
    ball
    0.87
    etooth
    0.86
     moist
    0.85
     quir
    0.85
    Activations Density 0.008%

    No Known Activations