© 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. 4-RES-JB
    11. 15289
    Prev
    Next
    INDEX
    Explanations

    phrases related to communication and discussion including phrases like "Hear me out."

    oai_token-act-pair · gpt-3.5-turbo

    expressions of emotional appeals and sentiments

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

    No Comments

    Negative Logits
     respectively
    -0.58
    inational
    -0.56
     afloat
    -0.48
    allery
    -0.48
     jointly
    -0.47
    qqa
    -0.47
    interstitial
    -0.47
    åĮ
    -0.47
     CAD
    -0.47
    è£ħ
    -0.47
    POSITIVE LOGITS
    Laughs
    0.56
    laughter
    0.55
     Rowling
    0.49
     Stranger
    0.48
     ESV
    0.48
    verb
    0.48
    laughs
    0.47
     Buster
    0.47
     Thief
    0.46
     Freak
    0.46
    Activations Density 2.135%

    No Known Activations