© 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. 11-RES-JB
    11. 24498
    Prev
    Next
    INDEX
    Explanations

    specific terms and phrases related to music, concepts, and datasets

    oai_token-act-pair · gpt-3.5-turbo

    references to specific events, concepts, or entities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     usefulness
    -0.70
     unnecess
    -0.70
     takeaway
    -0.69
     fallacy
    -0.69
     habits
    -0.68
     mindset
    -0.68
     yield
    -0.67
     limit
    -0.67
     underestimate
    -0.67
     wastes
    -0.67
    POSITIVE LOGITS
    named
    1.33
    accompanied
    1.31
    whose
    1.30
    which
    1.28
    designed
    1.24
    created
    1.21
    quartered
    1.17
    described
    1.17
    produced
    1.17
    published
    1.16
    Activations Density 0.279%

    No Known Activations