© 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. 8006
    Prev
    Next
    INDEX
    Explanations

    references to trouble or difficulties in various contexts

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "trouble."

    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
     Including
    -0.62
    omics
    -0.60
     based
    -0.60
    NI
    -0.58
     alloc
    -0.58
    Male
    -0.57
    rones
    -0.57
    Fram
    -0.56
    avi
    -0.56
    inav
    -0.56
    POSITIVE LOGITS
     trouble
    3.93
     troubles
    2.15
     Trouble
    2.01
     difficulty
    1.72
     difficulties
    1.69
     mischief
    1.62
     problems
    1.56
     danger
    1.46
     misfortune
    1.36
     problem
    1.35
    Activations Density 0.016%

    No Known Activations