INDEX

Explanations

political entities and government-related topics

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

aunder

-0.80

Ò

-0.76

tesy

-0.75

ãĤ´ãĥ³

-0.74

thia

-0.73

thood

-0.72

ictionary

-0.72

abetes

-0.70

CHAPTER

-0.70

Alert

-0.69

POSITIVE LOGITS

 slightest

1.13

 majority

1.13

 entire

1.01

 vast

0.95

 whole

0.94

 attackers

0.93

 average

0.93

 feds

0.93

 chances

0.92

 biggest

0.90

Activations Density 3.070%

political entities and government-related topics

No Comments

No Known Activations