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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 18925
    Prev
    Next
    INDEX
    Explanations

    changes in positions or job transitions

    oai_token-act-pair · gpt-3.5-turbo

    mentions of leadership transitions and official appointments

    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
    abiding
    -0.74
     smugglers
    -0.72
     postage
    -0.72
    banks
    -0.70
    models
    -0.69
     corrid
    -0.67
     spectators
    -0.65
    aband
    -0.65
    Supp
    -0.64
    violence
    -0.64
    POSITIVE LOGITS
     resign
    1.31
     resigned
    1.26
     resignation
    1.26
     replaces
    1.24
     appointed
    1.21
     appointment
    1.21
     replaced
    1.18
     appoint
    1.13
     successor
    1.11
     appointing
    1.10
    Activations Density 0.729%

    No Known Activations