INDEX

Explanations

words related to value and monetary measurements

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_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

jen

-0.87

waves

-0.68

éĹĺ

-0.67

ker

-0.65

bug

-0.65

ight

-0.63

Â¯Â¯Â¯Â¯

-0.60

Zen

-0.60

 Zimmer

-0.60

alis

-0.60

POSITIVE LOGITS

 proposition

1.14

 propositions

0.92

 value

0.91

 Proposition

0.90

uably

0.82

values

0.79

 Value

0.78

 values

0.73

uations

0.72

 judgments

0.72

Activations Density 0.629%

words related to value and monetary measurements

No Comments

No Known Activations