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. 0-RES-JB
    6. 5696
    Prev
    Next
    INDEX
    Explanations

    instances where something is imagined or envisioned

    oai_token-act-pair · gpt-3.5-turbo

    the concept of imagination or envisioning ideas and scenarios

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Quality
    -0.70
     turnover
    -0.70
     clearance
    -0.67
     Clear
    -0.66
     Strength
    -0.64
     Immunity
    -0.64
    Clear
    -0.64
    formance
    -0.63
     Citation
    -0.62
     Quality
    -0.62
    POSITIVE LOGITS
     imagined
    3.42
     envisioned
    2.55
     imagining
    2.23
     imag
    2.15
     dreamed
    2.14
     envis
    2.02
     envision
    1.95
     imaginary
    1.88
     imagine
    1.75
     fantas
    1.71
    Activations Density 0.027%

    No Known Activations