© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 0-RES-JB
    11. 10438
    Prev
    Next
    INDEX
    Explanations

    sentences that start with an exclamation or an interjection followed by a comma

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating strong emotional reactions or emphasis

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     requ
    -0.77
     gard
    -0.77
     brid
    -0.76
     favour
    -0.74
     harb
    -0.72
     seldom
    -0.71
     favoured
    -0.71
     enqu
    -0.71
     endeav
    -0.71
     mould
    -0.70
    POSITIVE LOGITS
    Okay
    2.46
     Okay
    2.30
    Alright
    2.20
     Seriously
    2.13
    Yep
    2.11
     Yep
    2.09
    Seriously
    1.99
     Nope
    1.98
     Alright
    1.97
     Hmm
    1.80
    Activations Density 0.049%

    No Known Activations