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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    3. GPT2-Small
    4. Residual Similar CE Local
    5. 6-RES_SCL-AJT
    6. 4177
    Prev
    Next
    INDEX
    Explanations

    economic and financial terms and related information

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scl-ajt/6-res_scl-ajt
    Prompts (Dashboard)
    12,288 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    46,080
    Data Type
    torch.float32
    Hook Point
    blocks.6.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    apollo-research/Skylion007-openwebtext-tokenizer-gpt2
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    447
    +0.11
    0.8%
    248
    +0.08
    0.6%
    508
    +0.07
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    679
    +0.11
    0.30
    248
    +0.08
    0.05
    640
    +0.07
    0.33
    Negative Logits
    xual
    -0.53
    othal
    -0.50
    drop
    -0.47
    ayn
    -0.46
    ĸļ
    -0.46
     subp
    -0.45
    arching
    -0.43
     coerc
    -0.42
    opian
    -0.42
    abies
    -0.41
    POSITIVE LOGITS
     headquartered
    0.50
    States
    0.44
     Sold
    0.43
    steen
    0.39
     Distribut
    0.39
     Bundy
    0.38
     Tycoon
    0.38
     Sales
    0.38
     Industry
    0.37
    Indust
    0.37
    Activations Density 25.877%

    No Known Activations