© 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. 8-RES-JB
    11. 17219
    Prev
    Next
    INDEX
    Explanations

    the presence of quotation marks

    oai_token-act-pair · gpt-3.5-turbo

    quotation marks and their surrounding phrases

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    •Partition Features Layer 8 GPT2 Small

    No Comments

    Negative Logits
     buoy
    -0.73
     tabloid
    -0.73
     arri
    -0.71
     seasoned
    -0.71
     Prim
    -0.71
     vigil
    -0.70
     differentiated
    -0.69
     disciplinary
    -0.69
     killers
    -0.69
     pacing
    -0.68
    POSITIVE LOGITS
    true
    1.32
    false
    1.30
    personal
    1.23
    yes
    1.22
    shall
    1.20
    normal
    1.19
    SELECT
    1.17
    cold
    1.17
    need
    1.16
    single
    1.15
    Activations Density 0.105%

    No Known Activations