© 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. 4-RES-JB
    11. 17114
    Prev
    Next
    INDEX
    Explanations

    terms related to increasing or challenging situations

    oai_token-act-pair · gpt-3.5-turbo

    terms related to increases and challenges

    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
     dust
    -0.75
     patiently
    -0.68
    meal
    -0.67
     directions
    -0.67
     grounds
    -0.65
     wipe
    -0.64
     walk
    -0.62
     brush
    -0.62
     lips
    -0.62
     belt
    -0.62
    POSITIVE LOGITS
    ating
    1.31
    ated
    1.30
    ational
    1.15
    ator
    1.14
    ators
    1.14
    aters
    1.11
    ates
    1.08
    at
    1.06
    ationally
    1.06
    ased
    1.05
    Activations Density 0.033%

    No Known Activations