© 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. 0-RES-JB
    11. 22200
    Prev
    Next
    INDEX
    Explanations

    phrases related to criminal activities such as robbery and harassment

    oai_token-act-pair · gpt-3.5-turbo

    instances of theft or victimization

    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
     appeal
    -0.79
     recommendation
    -0.75
     emphasis
    -0.74
     inference
    -0.73
     availability
    -0.73
     adoption
    -0.73
     mitigation
    -0.72
     deployment
    -0.72
     usage
    -0.72
     issuance
    -0.71
    POSITIVE LOGITS
     robbed
    2.07
     harassed
    2.02
     harmed
    1.94
     prosecuted
    1.82
     cheated
    1.82
     recruited
    1.79
     deceived
    1.79
     bullied
    1.78
     punished
    1.77
     interrogated
    1.76
    Activations Density 0.074%

    No Known Activations