© 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. 8-RES-JB
    11. 17034
    Prev
    Next
    INDEX
    Explanations

    proper nouns or specific terms repeated across different contexts

    oai_token-act-pair · gpt-3.5-turbo

    references to "third person" or "third party" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    staking
    -0.81
    veyard
    -0.70
    UTION
    -0.69
    immers
    -0.68
    ++++++++++++++++
    -0.68
    ITED
    -0.67
     Monsters
    -0.62
     Chow
    -0.60
     Limits
    -0.59
    Cache
    -0.59
    POSITIVE LOGITS
     baseman
    1.18
    rd
    1.12
     inning
    0.89
     consecutive
    0.86
    irty
    0.85
     installment
    0.81
    ousand
    0.80
     Reich
    0.79
    uous
    0.78
     quarter
    0.76
    Activations Density 0.032%

    No Known Activations