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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 20325
    Prev
    Next
    INDEX
    Explanations

    references to specific characters or concepts in a video game, likely related to gameplay strategies or character interactions

    oai_token-act-pair · gpt-3.5-turbo

    references to specific characters and their abilities in a gaming context

    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
    Not in Any Lists

    No Comments

    Negative Logits
    EngineDebug
    -0.89
     Flavoring
    -0.83
     Hubbard
    -0.83
     Toys
    -0.77
    ovie
    -0.77
     TSA
    -0.77
     Gutenberg
    -0.76
     VIDE
    -0.72
    utenberg
    -0.72
     Kubrick
    -0.72
    POSITIVE LOGITS
     aggro
    0.96
    lane
    0.90
     minions
    0.89
     jung
    0.86
     creeps
    0.84
     lane
    0.83
     ADC
    0.82
     minion
    0.82
    ande
    0.81
     meta
    0.81
    Activations Density 0.117%

    No Known Activations