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

    Neuronpedia

    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. 4-RES-JB
    11. 22337
    Prev
    Next
    INDEX
    Explanations

    phrases related to strength, power, dominance, and control

    oai_token-act-pair · gpt-3.5-turbo

    references to juggling and related activities

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

    No Comments

    Negative Logits
    */(
    -0.71
    REDACTED
    -0.65
     unconditional
    -0.65
     elevation
    -0.62
    WARD
    -0.62
    wards
    -0.61
     sentencing
    -0.61
     hospitality
    -0.60
     Conversation
    -0.60
     judgement
    -0.60
    POSITIVE LOGITS
    ernaut
    1.43
     jugg
    1.10
    glers
    1.04
     juggling
    0.96
    alos
    0.96
    alo
    0.95
    annon
    0.86
    ibles
    0.86
    adelphia
    0.84
    alore
    0.83
    Activations Density 0.012%

    No Known Activations