© 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. 4-RES-JB
    11. 15307
    Prev
    Next
    INDEX
    Explanations

    phrases related to world-building in a fictional setting

    oai_token-act-pair · gpt-3.5-turbo

    references to fictional worlds and their development

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 4-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    4
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     actionGroup
    -1.01
    etsk
    -0.91
    icer
    -0.89
    odox
    -0.88
    edes
    -0.76
    ratulations
    -0.76
    gotten
    -0.76
    ients
    -0.75
    ippi
    -0.70
    oug
    -0.70
    POSITIVE LOGITS
    building
    1.02
     Reincarn
    0.97
    wide
    0.88
    scape
    0.81
     Reborn
    0.78
     collide
    0.72
    ship
    0.71
    builder
    0.70
    arium
    0.70
     Creation
    0.70
    Activations Density 0.055%

    No Known Activations