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

    Neuronpedia

    Jacobian LensNEW
    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. 1950
    Prev
    Next
    INDEX
    Explanations

    phrases related to the act of observation or watching

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "observe" 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
     ton
    -0.71
     Honda
    -0.70
     pitched
    -0.68
     Hulk
    -0.67
     Klu
    -0.67
     Cadillac
    -0.66
     Punk
    -0.66
     Font
    -0.66
     Dropbox
    -0.63
     Duc
    -0.63
    POSITIVE LOGITS
     observe
    3.19
     observing
    2.94
     observ
    2.36
     observes
    1.97
     observer
    1.85
     observation
    1.70
    Obs
    1.49
     observations
    1.47
     Observ
    1.46
     observed
    1.43
    Activations Density 0.026%

    No Known Activations