INDEX

Explanations

mentions of geographical locations being affected in a negative way

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_fs6144-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_6144

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

6,144

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

sylv

-0.81

merce

-0.74

ancial

-0.74

escription

-0.73

SPONSORED

-0.70

oyer

-0.67

iquette

-0.66

olson

-0.65

JUST

-0.65

iveness

-0.64

POSITIVE LOGITS

 whence

1.11

 captivity

0.98

 nowhere

0.97

 obscurity

0.96

 anesthesia

0.83

 hiber

0.82

 hiding

0.81

 afar

0.80

 exile

0.76

 abroad

0.72

Activations Density 0.343%

mentions of geographical locations being affected in a negative way

No Comments

No Known Activations