© 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. 0-RES-JB
    11. 11726
    Prev
    Next
    INDEX
    Explanations

    phrases mentioning strength or intensity

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "strong" 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
     conversion
    -0.74
     journalism
    -0.70
     stuff
    -0.66
     archives
    -0.66
     utilities
    -0.64
     reimbursement
    -0.63
     elapsed
    -0.62
     tile
    -0.61
     conversions
    -0.61
     reimburse
    -0.61
    POSITIVE LOGITS
    strong
    4.19
    Strong
    1.95
    strength
    1.58
     Strong
    1.46
     strong
    1.43
    weak
    1.42
     strongest
    1.23
    Weak
    1.21
    tall
    1.15
     stronger
    1.14
    Activations Density 0.011%

    No Known Activations