© 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. 23329
    Prev
    Next
    INDEX
    Explanations

    strong emotional expressions and expletives

    oai_token-act-pair · gpt-3.5-turbo

    expressions of strong emotional reactions or frustrations

    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
    âĢij
    -1.10
    »
    -1.06
    âĢ
    -1.01
    âĢ¢
    -0.99
     ÂŃ
    -0.99
    ÂŃ
    -0.98
    "}],"
    -0.94
    Ó
    -0.89
    .","
    -0.89
     âĢ
    -0.87
    POSITIVE LOGITS
     shitty
    1.53
     fucking
    1.45
     fucked
    1.41
     bullshit
    1.41
     shit
    1.34
     FUCK
    1.34
     asshole
    1.33
     fuck
    1.31
     idiots
    1.30
     retarded
    1.28
    Activations Density 1.168%

    No Known Activations