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. 7249
    Prev
    Next
    INDEX
    Explanations

    organizations or teams involved in rescue operations

    oai_token-act-pair · gpt-3.5-turbo

    references to rescue organizations and wildlife conservation

    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
     peripheral
    -0.80
     lin
    -0.78
     imb
    -0.71
     productivity
    -0.71
     cord
    -0.70
     sidel
    -0.70
     col
    -0.69
     inert
    -0.68
     input
    -0.67
     commons
    -0.67
    POSITIVE LOGITS
     Rescue
    3.08
     Wildlife
    1.84
     Youth
    1.59
     Aid
    1.30
     Disaster
    1.29
     Rehab
    1.24
     Extreme
    1.23
     Revenge
    1.22
     Salvation
    1.17
     Asylum
    1.11
    Activations Density 0.050%

    No Known Activations