© 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. 4235
    Prev
    Next
    INDEX
    Explanations

    first sentences or phrases in text blocks

    oai_token-act-pair · gpt-3.5-turbo

    instances of statements and punctuation indicating opinions or mixed sentiments in a discussion

    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
     pocket
    -0.80
     magically
    -0.77
     stocking
    -0.74
     harvest
    -0.73
     handshake
    -0.71
     packing
    -0.71
     screw
    -0.71
     overhead
    -0.70
     swoop
    -0.70
     tricked
    -0.69
    POSITIVE LOGITS
     Critics
    1.33
     Questions
    1.25
    Questions
    1.24
    Critics
    1.18
     Some
    1.16
     Answers
    1.13
     Supporters
    1.10
     Several
    1.09
     Debate
    1.08
     Advoc
    1.05
    Activations Density 0.367%

    No Known Activations