© 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. 2-RES-JB
    11. 11148
    Prev
    Next
    INDEX
    Explanations

    phrases related to strong negative emotions or opinions

    oai_token-act-pair · gpt-3.5-turbo

    expressions of dislike or hatred towards various subjects

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    enture
    -0.88
    ItemImage
    -0.88
    SPONSORED
    -0.87
    interstitial
    -0.87
    aqu
    -0.86
    arta
    -0.86
    OGR
    -0.84
    aunder
    -0.77
    authent
    -0.76
    vernment
    -0.76
    POSITIVE LOGITS
     hated
    1.15
     hates
    0.93
     hate
    0.89
    fully
    0.85
    lihood
    0.82
    liest
    0.77
     Hate
    0.75
     revenge
    0.73
     hating
    0.72
     mistress
    0.71
    Activations Density 0.017%

    No Known Activations