© 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. 10119
    Prev
    Next
    INDEX
    Explanations

    similes comparing actions to physical force or intensity

    oai_token-act-pair · gpt-3.5-turbo

    similes and metaphors that employ the word "like."

    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
    VICE
    -0.82
    ulic
    -0.76
    inion
    -0.70
    inas
    -0.69
    icipated
    -0.68
    iets
    -0.67
     Together
    -0.67
    Enlarge
    -0.65
    SEE
    -0.65
    vantage
    -0.65
    POSITIVE LOGITS
     wildfire
    1.27
     crazy
    1.01
     clock
    0.93
    liest
    0.91
     mad
    0.89
    lier
    0.89
     bandits
    0.87
     rabbits
    0.82
     mushrooms
    0.82
     never
    0.81
    Activations Density 0.069%

    No Known Activations