© 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. 23101
    Prev
    Next
    INDEX
    Explanations

    phrases starting with "Being" describing various attributes or roles

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "Being" 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
     markets
    -0.62
    ga
    -0.60
    Äĩ
    -0.60
     pockets
    -0.59
    ops
    -0.58
    ansen
    -0.56
     seq
    -0.56
    utsch
    -0.56
     batches
    -0.56
     cleanup
    -0.55
    POSITIVE LOGITS
     Being
    3.29
    Being
    2.71
    being
    2.47
     being
    1.77
     Having
    1.42
     Knowing
    1.40
    Having
    1.32
     Saying
    1.30
     Seeing
    1.27
    having
    1.23
    Activations Density 0.009%

    No Known Activations