INDEX

Explanations

phrases related to advocating for change or improvement

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 pains

-0.71

tones

-0.67

stad

-0.66

 significance

-0.63

 particulars

-0.62

enhagen

-0.59

cern

-0.57

 floats

-0.57

udeau

-0.56

 plots

-0.56

POSITIVE LOGITS

through

0.83

 through

0.78

 probably

0.77

 undoubtedly

0.76

 indeed

0.74

 definitely

0.74

 simply

0.73

 ALWAYS

0.72

 surely

0.71

\\\\\\\\

0.70

Activations Density 0.076%

phrases related to advocating for change or improvement

No Comments

No Known Activations