© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 4-RES-JB
    11. 2899
    Prev
    Next
    INDEX
    Explanations

    the word "hug" or related actions and expressions

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "hug" and its variations

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

    No Comments

    Negative Logits
    piracy
    -0.76
    umer
    -0.68
    DoS
    -0.64
    YE
    -0.64
    protected
    -0.60
    ourses
    -0.60
    pir
    -0.58
    consumer
    -0.57
     piracy
    -0.57
    oder
    -0.57
    POSITIVE LOGITS
     hug
    0.99
     hugs
    0.96
    eness
    0.95
     Hug
    0.94
     goodbye
    0.91
    wrap
    0.90
     hugged
    0.88
     hugging
    0.84
    eson
    0.82
    nesday
    0.82
    Activations Density 0.007%

    No Known Activations