© 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. 11-RES-JB
    11. 10027
    Prev
    Next
    INDEX
    Explanations

    specific clauses with restrictive language or imposing limitations

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to policies and regulations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    noticed
    -0.74
     Horizons
    -0.68
     Ready
    -0.66
    odiac
    -0.66
     Grip
    -0.64
    DragonMagazine
    -0.64
    Ready
    -0.64
    GGGGGGGG
    -0.63
    nice
    -0.63
     Jol
    -0.63
    POSITIVE LOGITS
     violate
    1.46
     jeopard
    1.41
     undermine
    1.41
     infring
    1.39
     endanger
    1.39
     impede
    1.31
     unnecessarily
    1.26
     adversely
    1.25
     deprive
    1.24
     harm
    1.23
    Activations Density 0.322%

    No Known Activations