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

    Neuronpedia

    APICircuit 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. 21322
    Prev
    Next
    INDEX
    Explanations

    verbs related to growth, creation, or advancement

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "develop" and its variations

    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
     Coulter
    -0.69
     Faw
    -0.68
     prayers
    -0.66
     Prayer
    -0.64
     Bucket
    -0.63
    iture
    -0.62
     condol
    -0.62
    Flags
    -0.62
    resy
    -0.60
    tons
    -0.59
    POSITIVE LOGITS
     develop
    3.21
     develops
    2.50
     developing
    2.21
    develop
    2.17
     Develop
    2.09
     developed
    2.07
    Develop
    2.06
     development
    1.85
    developed
    1.73
     evolve
    1.69
    Activations Density 0.017%

    No Known Activations