© 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. 10-RES-JB
    11. 17387
    Prev
    Next
    INDEX
    Explanations

    news headlines or stories related to legal actions and sentences

    oai_token-act-pair · gpt-3.5-turbo

    references to criminal cases and legal outcomes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    colored
    -0.91
     favors
    -0.86
     colorful
    -0.85
     honor
    -0.82
     rumors
    -0.82
     behaviors
    -0.80
     honored
    -0.79
     traveled
    -0.79
     honorable
    -0.78
     labor
    -0.78
    POSITIVE LOGITS
     whilst
    1.53
     (£
    1.45
     £
    1.45
     Whilst
    1.44
     Belfast
    1.34
     recognise
    1.33
    '.
    1.29
     recognised
    1.28
     programme
    1.24
     organising
    1.23
    Activations Density 1.127%

    No Known Activations