INDEX

Explanations

words related to government, policy, and economics

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

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

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ãĥİ

-0.88

bury

-0.76

 Photograph

-0.74

mt

-0.70

mol

-0.69

scribe

-0.69

folk

-0.68

window

-0.67

ija

-0.66

ÃŃa

-0.65

POSITIVE LOGITS

 sarcast

0.92

 noting

0.90

 bluntly

0.78

 referring

0.78

 anecdotes

0.77

 rhet

0.77

 pointing

0.77

 examples

0.76

 emphas

0.75

 scathing

0.75

Activations Density 0.265%

words related to government, policy, and economics

No Comments

No Known Activations