© 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. 16476
    Prev
    Next
    INDEX
    Explanations

    words related to retaliation or revenge

    oai_token-act-pair · gpt-3.5-turbo
    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
    leaf
    -0.67
     Fold
    -0.63
     Clin
    -0.61
    oscope
    -0.61
     fold
    -0.61
     electron
    -0.60
    Found
    -0.60
    Station
    -0.58
    cyl
    -0.58
     Canal
    -0.58
    POSITIVE LOGITS
     retaliation
    3.97
     retali
    3.15
     retribution
    2.87
     retaliate
    2.81
     repr
    2.31
     revenge
    2.10
    aliation
    2.01
     vengeance
    1.65
    venge
    1.46
     punitive
    1.35
    Activations Density 0.023%

    No Known Activations