© 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. 1-RES-JB
    11. 2596
    Prev
    Next
    INDEX
    Explanations

    phrases related to physical movement or activity

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "shake" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    etheless
    -0.78
    minist
    -0.72
    atar
    -0.69
    inately
    -0.67
    uct
    -0.67
     Defenders
    -0.65
    efficient
    -0.65
     [+
    -0.64
     Kinnikuman
    -0.64
     Ducks
    -0.63
    POSITIVE LOGITS
     shake
    1.28
     shakes
    1.01
     Shake
    1.01
     shaking
    0.91
    alore
    0.83
    shake
    0.81
     shaken
    0.81
    stals
    0.80
    borough
    0.78
    undown
    0.78
    Activations Density 0.012%

    No Known Activations