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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 7454
    Prev
    Next
    INDEX
    Explanations

    programming code keywords and related terms

    oai_token-act-pair · gpt-3.5-turbo

    references to coding and programming within text

    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
     wet
    -0.72
     Peterson
    -0.71
     mounts
    -0.68
     rolling
    -0.66
     Russ
    -0.66
     soaking
    -0.64
     roller
    -0.64
     aging
    -0.63
     viewpoint
    -0.63
     embell
    -0.63
    POSITIVE LOGITS
    Code
    4.14
    code
    2.54
     Code
    2.36
    codes
    2.23
     CODE
    2.18
     code
    2.02
     Codes
    1.90
     codes
    1.88
    coded
    1.53
     coded
    1.47
    Activations Density 0.010%

    No Known Activations