© 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. 10-RES-JB
    11. 24397
    Prev
    Next
    INDEX
    Explanations

    pairs of sentences where the second one continues from where the first one left off, creating a coherent paragraph

    oai_token-act-pair · gpt-3.5-turbo

    instances of violence or trauma

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     enriched
    -0.98
     innov
    -0.91
     persu
    -0.91
     shrinking
    -0.91
     persuasion
    -0.90
     advant
    -0.89
     quir
    -0.85
     oppressed
    -0.84
     striving
    -0.84
     rebell
    -0.84
    POSITIVE LOGITS
    Investigators
    2.04
    Police
    2.01
    Authorities
    1.95
    Officers
    1.90
    Detect
    1.77
    Officials
    1.74
    Witness
    1.72
    According
    1.69
    Prosecutors
    1.67
    Neigh
    1.64
    Activations Density 0.293%

    No Known Activations