© 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. 11013
    Prev
    Next
    INDEX
    Explanations

    phrases indicating something is wrong or needs attention

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating a sense of something being wrong or missing

    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
    stood
    -0.67
     fame
    -0.65
     resume
    -0.63
    vale
    -0.61
     suburb
    -0.59
     Span
    -0.59
     Rated
    -0.58
     Documents
    -0.58
     careers
    -0.57
     examples
    -0.57
    POSITIVE LOGITS
     wrong
    1.19
    wrong
    1.00
     happening
    0.99
     missing
    0.94
     bothering
    0.93
     rotten
    0.92
     horribly
    0.90
     terribly
    0.86
    missing
    0.83
     strang
    0.79
    Activations Density 0.088%

    No Known Activations