© 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. 16884
    Prev
    Next
    INDEX
    Explanations

    phrases related to the concept of reversing a decision or action

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "reverse" and its variations, indicating changes or alterations

    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
    Ball
    -0.72
    camp
    -0.68
    bies
    -0.67
    esp
    -0.65
    piring
    -0.65
    bat
    -0.65
    illa
    -0.65
    bel
    -0.64
    bag
    -0.63
    ager
    -0.63
    POSITIVE LOGITS
     reversed
    3.87
     reversing
    2.33
     reversal
    2.23
     revers
    2.23
     reverse
    2.15
    reverse
    1.94
     reversible
    1.81
     overturned
    1.62
     inverted
    1.62
     Reverse
    1.53
    Activations Density 0.016%

    No Known Activations