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. 20992
    Prev
    Next
    INDEX
    Explanations

    words related to achieving goals or completing tasks effectively

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "accomplish" and related discussions of achievement

    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
     UK
    -0.79
     £
    -0.74
     Leigh
    -0.72
     centres
    -0.70
    UK
    -0.70
     scept
    -0.68
     NHS
    -0.65
     Queensland
    -0.64
     centre
    -0.63
     NSW
    -0.62
    POSITIVE LOGITS
     accomplish
    3.16
     achieve
    1.93
     accompl
    1.63
     accomplished
    1.56
     achieves
    1.41
     accomplishment
    1.37
     fulfill
    1.37
     succeed
    1.32
     Achieve
    1.30
     attain
    1.27
    Activations Density 0.013%

    No Known Activations