INDEX

Explanations

phrases indicating "in other words."

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

apego

-0.67

yip

-0.66

atism

-0.64

 overcame

-0.61

avorite

-0.59

 outweigh

-0.58

icides

-0.58

Always

-0.57

iste

-0.56

achelor

-0.56

POSITIVE LOGITS

words

1.11

worldly

1.06

 words

1.04

 respects

0.94

wise

0.86

 contexts

0.85

word

0.83

 instances

0.80

 circumstances

0.78

 areas

0.77

Activations Density 0.016%

phrases indicating "in other words."

No Comments

No Known Activations