© 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. 3-RES-JB
    11. 10664
    Prev
    Next
    INDEX
    Explanations

    words related to verbal communication, particularly instances of verbal abuse

    oai_token-act-pair · gpt-3.5-turbo

    instances of verbal communication or abuse

    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
    akeru
    -0.81
    annis
    -0.78
    resses
    -0.77
    uden
    -0.76
    enthal
    -0.75
    ocrates
    -0.74
    ressing
    -0.73
    roxy
    -0.72
    ramid
    -0.71
    arent
    -0.71
    POSITIVE LOGITS
     verbal
    0.98
    isations
    0.95
    ized
    0.93
     altercation
    0.91
    ization
    0.86
    izations
    0.86
     cues
    0.85
    ãĥ£
    0.82
     communication
    0.82
     representations
    0.80
    Activations Density 0.008%

    No Known Activations