© 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. 17909
    Prev
    Next
    INDEX
    Explanations

    words related to threats or threats of harm

    oai_token-act-pair · gpt-3.5-turbo

    references to the word "threatened" and its variations, indicating instances of danger or risk

    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
     Bounce
    -0.66
     Alloy
    -0.65
    igm
    -0.65
    RT
    -0.65
     Ded
    -0.63
    aneous
    -0.61
    aus
    -0.60
     Doodle
    -0.59
     Omni
    -0.58
    Birth
    -0.58
    POSITIVE LOGITS
     threatened
    3.61
     threaten
    2.18
    threat
    2.16
     threatening
    2.03
     threatens
    2.02
     endangered
    1.91
     threats
    1.78
     Threat
    1.75
     threat
    1.63
     harassed
    1.54
    Activations Density 0.017%

    No Known Activations