© 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. 6181
    Prev
    Next
    INDEX
    Explanations

    instances where the concept of 'never' is referenced in context

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "never."

    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
    ahime
    -0.90
    pour
    -0.87
     antioxid
    -0.78
    iant
    -0.71
    åĤ
    -0.67
    achev
    -0.67
    CRIP
    -0.63
    uliffe
    -0.63
    PI
    -0.63
    ESCO
    -0.61
    POSITIVE LOGITS
    theless
    1.74
     bothered
    1.12
     dreamed
    0.99
     ceases
    0.95
     existed
    0.85
     ceased
    0.85
    mind
    0.84
     heard
    0.84
     bothers
    0.82
     imagined
    0.81
    Activations Density 0.059%

    No Known Activations