© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 21504
    Prev
    Next
    INDEX
    Explanations

    mentions of different states or state departments with particular focus on activism or law enforcement

    oai_token-act-pair · gpt-3.5-turbo

    references to government bodies or political entities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    kens
    -0.72
    fig
    -0.67
    ĸļ
    -0.66
    scar
    -0.62
    XXXX
    -0.61
    helm
    -0.60
     scatter
    -0.59
    irsch
    -0.59
     Zup
    -0.59
    cdn
    -0.59
    POSITIVE LOGITS
     State
    3.75
    State
    2.83
     STATE
    2.79
     state
    2.19
    state
    2.17
    STATE
    2.08
     States
    1.84
    States
    1.70
     Commonwealth
    1.51
     Federal
    1.32
    Activations Density 0.029%

    No Known Activations