Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 7-RES-JB
    6. 6304
    Prev
    Next
    INDEX
    Explanations

    expressions of regret or missed opportunities

    oai_token-act-pair · gpt-3.5-turbo

    phrases involving the conditional "should have."

    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
     hostage
    -0.66
     maze
    -0.63
     dement
    -0.61
     Wid
    -0.60
    ocol
    -0.59
     capsule
    -0.59
    Fs
    -0.58
     muse
    -0.57
    Mem
    -0.57
     enchant
    -0.57
    POSITIVE LOGITS
     gotten
    1.06
    been
    1.02
     been
    1.00
    gotten
    0.93
     listened
    0.90
     behaved
    0.88
     avoided
    0.84
     waited
    0.84
     consulted
    0.82
     existed
    0.81
    Activations Density 0.041%

    No Known Activations