© 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. 16830
    Prev
    Next
    INDEX
    Explanations

    instances of the word "save" encoded with strong emphasis

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "save"

    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
     Lia
    -0.72
    argon
    -0.68
    mr
    -0.66
    rum
    -0.66
     Minority
    -0.66
     LAR
    -0.65
    eele
    -0.64
     Joint
    -0.62
    rus
    -0.61
    hibited
    -0.61
    POSITIVE LOGITS
     save
    3.87
    save
    2.93
     saves
    2.76
     Save
    2.43
     saving
    2.26
     saved
    2.23
    Save
    2.20
     Saving
    1.53
    saving
    1.53
     conserve
    1.32
    Activations Density 0.021%

    No Known Activations