© 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. 6220
    Prev
    Next
    INDEX
    Explanations

    phrases related to admission or confession

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "mit" and its variations, indicating a focus on expressions related to permission or admission

    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
     scoop
    -0.76
     editions
    -0.71
     wra
    -0.63
     titles
    -0.62
     cloth
    -0.62
     sizes
    -0.62
     rotated
    -0.62
     roaming
    -0.61
     Clubs
    -0.61
     boxes
    -0.61
    POSITIVE LOGITS
    mit
    4.75
    mits
    2.64
    MIT
    2.54
    mitting
    2.04
    mitted
    1.84
    mitter
    1.79
    Mit
    1.60
    mission
    1.58
    mite
    1.50
    mitt
    1.50
    Activations Density 0.007%

    No Known Activations