© 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. 6628
    Prev
    Next
    INDEX
    Explanations

    sentences related to societal or industrial observations

    oai_token-act-pair · gpt-3.5-turbo

    statements related to the state of various industries or societal conditions

    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
     Increases
    -0.70
     knocks
    -0.69
     Signs
    -0.69
     Benefits
    -0.65
     Tests
    -0.65
     Deaths
    -0.65
     caches
    -0.64
     warranties
    -0.64
    umatic
    -0.64
     Stones
    -0.63
    POSITIVE LOGITS
     comprised
    1.44
     composed
    1.30
     dominated
    1.18
     divided
    1.15
     undergoing
    1.11
     fragmented
    1.06
     united
    1.05
     populated
    1.05
     founded
    1.03
     reeling
    1.03
    Activations Density 0.302%

    No Known Activations