© 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 EvalsExportsinterp-engineNEWAPI 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. 8-RES-JB
    11. 11336
    Prev
    Next
    INDEX
    Explanations

    words related to routines or daily activities

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to routine activities and conditions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    urches
    -0.85
    rooms
    -0.79
    assies
    -0.78
    akings
    -0.76
    reddits
    -0.74
     Machines
    -0.74
    rencies
    -0.73
    blems
    -0.72
    ernels
    -0.72
    casts
    -0.72
    POSITIVE LOGITS
     standpoint
    1.04
     regimen
    0.77
     basis
    0.77
     perspective
    0.73
     approximation
    0.68
    etermined
    0.68
     handful
    0.68
     opponent
    0.67
     colleague
    0.65
     breeze
    0.65
    Activations Density 0.789%

    No Known Activations