© 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. 7-RES-JB
    11. 11832
    Prev
    Next
    INDEX
    Explanations

    rare, significant, and impactful adjectives or phrases

    oai_token-act-pair · gpt-3.5-turbo

    words and phrases that convey characteristics or conditions associated with rarity, complexity, and distinctiveness

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 7-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    7
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    alion
    -0.81
    achu
    -0.80
    ionage
    -0.73
    andom
    -0.73
    EMA
    -0.73
    APTER
    -0.73
    ACY
    -0.72
    ttp
    -0.72
    ulhu
    -0.72
    rower
    -0.71
    POSITIVE LOGITS
     enough
    1.01
     themselves
    0.96
     extensions
    0.96
     reminders
    0.93
     additions
    0.92
     substitutes
    0.87
     favourites
    0.86
     expressions
    0.86
     representations
    0.85
     creations
    0.84
    Activations Density 0.457%

    No Known Activations