© 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. 4-RES-JB
    11. 374
    Prev
    Next
    INDEX
    Explanations

    words or phrases related to opposition or conflict

    oai_token-act-pair · gpt-3.5-turbo

    references to anti-related sentiments or movements

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

    No Comments

    Negative Logits
    ulhu
    -0.93
    swick
    -0.82
    çīĪ
    -0.74
    ADRA
    -0.72
    ynski
    -0.71
     WD
    -0.70
    ĸļ
    -0.67
    ħĭ
    -0.66
     Cohn
    -0.65
     Sponsor
    -0.65
    POSITIVE LOGITS
    government
    1.01
    establishment
    0.99
    social
    0.98
    violence
    0.95
    Semitic
    0.93
    commun
    0.92
    American
    0.91
    democratic
    0.90
    capitalist
    0.89
    aligned
    0.89
    Activations Density 0.048%

    No Known Activations