© 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. 6864
    Prev
    Next
    INDEX
    Explanations

    phrases indicating exceptions or contrasts

    oai_token-act-pair · gpt-3.5-turbo

    the word "Except" and its variations, indicating exclusions or exceptions in statements

    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
     quickly
    -0.64
    uli
    -0.60
     leaked
    -0.59
     microbiome
    -0.57
    wen
    -0.55
     crowd
    -0.55
    elsen
    -0.55
     spread
    -0.55
     fellow
    -0.54
     recently
    -0.54
    POSITIVE LOGITS
     Except
    3.48
    Except
    3.07
    except
    2.68
     except
    1.66
     Unless
    1.47
    Unless
    1.33
     Notwithstanding
    1.32
    unless
    1.14
     Including
    1.13
    excluding
    1.12
    Activations Density 0.025%

    No Known Activations