© 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. 6838
    Prev
    Next
    INDEX
    Explanations

    terms related to mental processes or decision-making

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "think" and its variations, indicating a focus on thoughts or contemplation

    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
     cumulative
    -0.74
     total
    -0.72
     intact
    -0.69
     relief
    -0.69
     Completed
    -0.68
     recovery
    -0.66
     confirmed
    -0.65
     =================================================================
    -0.65
     payable
    -0.64
     remaining
    -0.63
    POSITIVE LOGITS
    think
    4.36
    thinking
    2.04
    thought
    1.94
    Think
    1.79
    fuck
    1.32
    speak
    1.29
     thinker
    1.17
     Think
    1.17
    say
    1.16
    meaning
    1.15
    Activations Density 0.009%

    No Known Activations