© 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. 6785
    Prev
    Next
    INDEX
    Explanations

    the word "sees" followed by an object or action

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "sees"

    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
    ongh
    -0.65
    alsa
    -0.56
     simple
    -0.54
     Oklahoma
    -0.54
     Miscellaneous
    -0.54
    ab
    -0.53
    umber
    -0.53
    restricted
    -0.53
    umo
    -0.52
    wordpress
    -0.51
    POSITIVE LOGITS
     sees
    3.23
     saw
    1.93
     finds
    1.58
     hears
    1.57
     recognizes
    1.56
     see
    1.56
     thinks
    1.54
     expects
    1.52
     realizes
    1.48
     deems
    1.47
    Activations Density 0.010%

    No Known Activations