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. 3-RES-JB
    6. 2169
    Prev
    Next
    INDEX
    Explanations

    phrases related to strong emotions or opinions expressed by individuals or groups

    oai_token-act-pair · gpt-3.5-turbo

    phrases expressing strong concern or emotional intensity

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    iful
    -0.78
     Millennium
    -0.73
    arters
    -0.73
    amins
    -0.71
    ulia
    -0.70
    othy
    -0.70
    ordan
    -0.70
    ifully
    -0.66
     Athlet
    -0.66
    ioch
    -0.65
    POSITIVE LOGITS
     ingrained
    0.96
     immersed
    0.86
     rooted
    0.85
     embedded
    0.84
     enough
    0.83
     engaged
    0.81
     seeded
    0.81
     eng
    0.80
     connected
    0.80
     entrenched
    0.80
    Activations Density 0.010%

    No Known Activations