Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 1-RES-JB
    6. 9211
    Prev
    Next
    INDEX
    Explanations

    words related to physical pressure or force

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "pressed" 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
    eway
    -0.64
    flame
    -0.62
    ordinate
    -0.61
    Ark
    -0.61
     Jet
    -0.60
    liest
    -0.59
    mite
    -0.59
    fin
    -0.58
     Rays
    -0.58
     Pipeline
    -0.57
    POSITIVE LOGITS
     pressed
    1.39
    pressed
    1.25
     presses
    0.96
    press
    0.95
    ierrez
    0.91
    ername
    0.90
     showc
    0.88
     PRESS
    0.88
     pressing
    0.87
    ivated
    0.81
    Activations Density 0.004%

    No Known Activations