© 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. 15811
    Prev
    Next
    INDEX
    Explanations

    references to dehydration

    oai_token-act-pair · gpt-3.5-turbo

    references to dehydration

    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
     maiden
    -0.76
     Quinn
    -0.70
    lees
    -0.70
     Ops
    -0.69
     Stra
    -0.69
     Equity
    -0.68
    brid
    -0.67
    lein
    -0.67
    Stra
    -0.66
     Lever
    -0.66
    POSITIVE LOGITS
     dehyd
    3.61
     dehydration
    2.26
     incub
    1.46
     humid
    1.36
     humidity
    1.27
     Sere
    1.25
     drying
    1.22
     evapor
    1.17
     thirsty
    1.14
     deterior
    1.03
    Activations Density 0.055%

    No Known Activations