© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 0-RES-JB
    11. 18043
    Prev
    Next
    INDEX
    Explanations

    words related to investigations

    oai_token-act-pair · gpt-3.5-turbo

    mentions of investigations

    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
     Signed
    -0.72
    jar
    -0.70
     Tradable
    -0.70
    malink
    -0.68
     distribut
    -0.64
    ↵Âł
    -0.63
     tee
    -0.63
    minus
    -0.63
     NAME
    -0.62
    atum
    -0.61
    POSITIVE LOGITS
     investigations
    3.46
     Investigations
    2.36
     investigation
    2.20
     inquiries
    2.07
     probes
    1.99
     prosecutions
    1.98
     investigators
    1.85
     Investigation
    1.81
     examinations
    1.74
     investigative
    1.72
    Activations Density 0.014%

    No Known Activations