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

    Neuronpedia

    APICircuit 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. 6-RES-JB
    6. 1148
    Prev
    Next
    INDEX
    Explanations

    contractions of the form "verb + 's", typically indicating possession or omission of a letter

    oai_token-act-pair · gpt-3.5-turbo

    possessive constructions and questions about different subjects or topics

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 6-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    velop
    -0.70
    oak
    -0.62
    igraph
    -0.62
    onent
    -0.60
    sha
    -0.60
    UME
    -0.59
    erers
    -0.58
    cember
    -0.57
     wards
    -0.57
    onz
    -0.56
    POSITIVE LOGITS
     happened
    0.97
     happening
    0.88
     gonna
    0.87
     transpired
    0.84
    pace
    0.79
     happ
    0.75
     gotta
    0.73
     Done
    0.71
     REALLY
    0.68
     done
    0.68
    Activations Density 0.047%

    No Known Activations