© 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. 0-RES-JB
    11. 13913
    Prev
    Next
    INDEX
    Explanations

    words related to moisture or wetness

    oai_token-act-pair · gpt-3.5-turbo

    terms related to moisture and decay

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    avis
    -0.79
     Objective
    -0.69
     Gen
    -0.68
    aurus
    -0.68
    eye
    -0.67
    sch
    -0.66
     Chic
    -0.64
     Afric
    -0.64
    zynski
    -0.63
    boro
    -0.63
    POSITIVE LOGITS
     damp
    3.83
     moist
    1.20
     drying
    1.08
     soaked
    1.07
     atten
    1.03
     cush
    1.00
     stiffness
    1.00
    amping
    0.99
     soaking
    0.98
     mould
    0.98
    Activations Density 0.032%

    No Known Activations