© 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. 2784
    Prev
    Next
    INDEX
    Explanations

    comparisons of increasing strength or intensity

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "stronger" along with relevant contextual discussions about strength or improvement

    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
    opsis
    -0.67
     Topics
    -0.64
    iasco
    -0.63
    ania
    -0.62
    200000
    -0.61
     Waste
    -0.60
     checkout
    -0.59
     describes
    -0.59
    oad
    -0.59
     Minutes
    -0.59
    POSITIVE LOGITS
     stronger
    3.52
     weaker
    2.72
     strongest
    2.13
     strong
    1.95
     strengthened
    1.92
     tougher
    1.91
     sharper
    1.91
     weakest
    1.74
     weakened
    1.72
     thicker
    1.72
    Activations Density 0.013%

    No Known Activations