© 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. 11091
    Prev
    Next
    INDEX
    Explanations

    words related to lies and deception

    oai_token-act-pair · gpt-3.5-turbo

    references to deception or false information, particularly in relation to medical, environmental, and social issues

    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
    adelphia
    -0.92
     Mobility
    -0.90
    ktop
    -0.83
     Schwar
    -0.82
     Ports
    -0.82
    allery
    -0.77
    Rail
    -0.74
    ahime
    -0.74
     corridor
    -0.74
    esa
    -0.74
    POSITIVE LOGITS
     falsehood
    1.76
     debunk
    1.63
     hoax
    1.58
     misinformation
    1.53
     factual
    1.52
     false
    1.51
     disinformation
    1.51
    truth
    1.50
     truthful
    1.49
    False
    1.49
    Activations Density 0.899%

    No Known Activations