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. 1-RES-JB
    6. 19505
    Prev
    Next
    INDEX
    Explanations

    the phrase "two-thirds" indicating a specific proportion

    oai_token-act-pair · gpt-3.5-turbo

    mentions of fractions, specifically two-thirds

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Trend
    -0.74
     pastoral
    -0.64
     metab
    -0.60
    talk
    -0.60
    Journal
    -0.59
     Narr
    -0.57
    birds
    -0.57
     anx
    -0.56
    mon
    -0.55
    verbs
    -0.55
    POSITIVE LOGITS
    thirds
    1.17
     ILCS
    0.98
    quarters
    0.89
    avorite
    0.89
     thereof
    0.88
    cence
    0.87
    enary
    0.84
    fif
    0.79
    rils
    0.79
     "$:/
    0.78
    Activations Density 0.008%

    No Known Activations