© 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. 0-RES-JB
    11. 20353
    Prev
    Next
    INDEX
    Explanations

    phrases related to strong criticism or negative judgment

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "slam" and its variations in various contexts

    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
     Advent
    -0.73
     Provision
    -0.69
    eret
    -0.67
     Volunteers
    -0.64
    Tea
    -0.64
     Wand
    -0.62
    onge
    -0.62
     acquisition
    -0.61
    chie
    -0.60
    maps
    -0.60
    POSITIVE LOGITS
     slammed
    3.31
     slams
    2.89
     slamming
    2.77
     slam
    2.70
     Slam
    1.75
     blasted
    1.66
     smashed
    1.41
     lashed
    1.39
     slapped
    1.34
     pounded
    1.33
    Activations Density 0.024%

    No Known Activations