Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 11-RES-JB
    6. 15249
    Prev
    Next
    INDEX
    Explanations

    emotional states and reactions such as dismay, regret, alarm, concern, shock, anger, disappointment, and outrage expressed by individuals

    oai_token-act-pair · gpt-3.5-turbo

    expressions of strong emotions, particularly negative feelings such as dismay, outrage, and concern

    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
    eg
    -0.72
    icle
    -0.61
    catentry
    -0.60
    nice
    -0.60
     sag
    -0.58
    åī
    -0.58
    eor
    -0.58
    quickShipAvailable
    -0.57
    asus
    -0.56
    aband
    -0.56
    POSITIVE LOGITS
     over
    1.19
     regarding
    1.03
     about
    1.03
     ABOUT
    0.86
     concerning
    0.82
     towards
    0.82
    ously
    0.82
    lication
    0.81
     that
    0.80
     toward
    0.79
    Activations Density 0.107%

    No Known Activations