© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 3-RES-JB
    6. 23236
    Prev
    Next
    INDEX
    Explanations

    phrases indicating criminal charges or offenses

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to criminal charges and legal actions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     partName
    -0.88
    phrine
    -0.85
    Tokens
    -0.78
    ersed
    -0.77
    cms
    -0.76
    marg
    -0.75
    soType
    -0.72
     endif
    -0.71
    Enabled
    -0.70
    readable
    -0.69
    POSITIVE LOGITS
     Prosecut
    0.90
     Sergeant
    0.86
     manslaughter
    0.83
     perjury
    0.82
     homicide
    0.81
     Investigator
    0.79
     treason
    0.78
     sergeant
    0.78
     prosecution
    0.78
     Starr
    0.78
    Activations Density 0.238%

    No Known Activations