© 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. 10-RES-JB
    11. 14168
    Prev
    Next
    INDEX
    Explanations

    words related to confessions or admissions of guilt or wrongdoing

    oai_token-act-pair · gpt-3.5-turbo

    expressions of acknowledgment or admission of wrongdoing

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    asus
    -0.84
     ILCS
    -0.76
    assic
    -0.74
    rouse
    -0.74
    osi
    -0.72
    phabet
    -0.72
    DragonMagazine
    -0.71
    rior
    -0.71
    ighth
    -0.70
    prev
    -0.70
    POSITIVE LOGITS
     defeat
    1.20
     wrongdoing
    1.16
     guilt
    1.08
     mistakes
    1.00
     ignorance
    0.93
     fault
    0.92
     responsibility
    0.90
     shortcomings
    0.88
     admitting
    0.87
     failings
    0.84
    Activations Density 0.051%

    No Known Activations