Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 11-RES-JB
    6. 9604
    Prev
    Next
    INDEX
    Explanations

    phrases indicating permission or invitation

    oai_token-act-pair · gpt-3.5-turbo

    phrases encouraging freedom and the invitation to share opinions

    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
    fam
    -0.67
    ELD
    -0.65
    ritical
    -0.63
    dated
    -0.61
    quickShipAvailable
    -0.60
    croft
    -0.60
    lehem
    -0.59
    assium
    -0.59
    iron
    -0.59
     Adin
    -0.59
    POSITIVE LOGITS
     choose
    1.14
     roam
    1.10
     decide
    1.10
     improv
    1.05
     experiment
    1.01
     pursue
    1.00
     criticize
    0.98
     explore
    0.98
     disagree
    0.96
     modify
    0.95
    Activations Density 0.115%

    No Known Activations