© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 8037
    Prev
    Next
    INDEX
    Explanations

    verbs related to applying pressure or tightness

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "squeeze" and its variations

    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
     delegates
    -0.76
     Hasan
    -0.73
     monuments
    -0.69
     Tribal
    -0.69
    nom
    -0.69
    Cons
    -0.65
     pursuant
    -0.64
     Proto
    -0.63
    cons
    -0.62
    acons
    -0.62
    POSITIVE LOGITS
     squeeze
    3.69
     squeezed
    2.32
     squeezing
    2.31
     squee
    2.10
     pinch
    1.94
     cram
    1.30
     shake
    1.20
     shove
    1.12
     strain
    1.12
     shave
    1.10
    Activations Density 0.011%

    No Known Activations