© 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. 1-RES-JB
    11. 3249
    Prev
    Next
    INDEX
    Explanations

    specific days of the week, particularly Fridays

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "Friday" 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
    ership
    -0.83
    eas
    -0.79
    arist
    -0.77
    abet
    -0.77
    umbn
    -0.74
    ection
    -0.69
    hip
    -0.68
    ris
    -0.68
    abetic
    -0.68
    haps
    -0.68
    POSITIVE LOGITS
     morning
    1.37
     afternoon
    1.36
     mornings
    1.29
     night
    1.16
     evening
    1.14
     nights
    1.12
     Friday
    1.08
     Night
    1.07
     evenings
    0.98
     Thursday
    0.94
    Activations Density 0.018%

    No Known Activations