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. 8-RES-JB
    6. 20829
    Prev
    Next
    INDEX
    Explanations

    phrases related to video game titles

    oai_token-act-pair · gpt-3.5-turbo
    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
     parity
    -0.79
     FAA
    -0.76
     behavi
    -0.75
     former
    -0.74
     privacy
    -0.74
     telecommunications
    -0.74
     taxpayer
    -0.73
     analyst
    -0.72
     senior
    -0.72
     ÂŃ
    -0.72
    POSITIVE LOGITS
    RPG
    1.01
     Nebula
    1.00
    Sword
    0.99
    opl
    0.94
    sword
    0.94
    Evil
    0.93
    udos
    0.93
    opter
    0.92
    oreal
    0.92
    odiac
    0.91
    Activations Density 0.477%

    No Known Activations