© 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. 8-RES-JB
    11. 20239
    Prev
    Next
    INDEX
    Explanations

    language related to fabricated or misleading information, possibly in the context of politics or media

    oai_token-act-pair · gpt-3.5-turbo

    discussions of manipulation or distortion of truth

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    DragonMagazine
    -0.66
    wana
    -0.65
     âĨij
    -0.63
     Originally
    -0.62
    adobe
    -0.61
    trak
    -0.61
     Aberdeen
    -0.59
    raltar
    -0.58
    ursday
    -0.58
     Warehouse
    -0.58
    POSITIVE LOGITS
     predetermined
    0.79
     inevitably
    0.72
     meaningless
    0.72
     truths
    0.71
     insur
    0.70
     disson
    0.69
     trivial
    0.69
     downright
    0.69
     irresist
    0.68
     subconscious
    0.68
    Activations Density 1.139%

    No Known Activations