INDEX

Explanations

contractions

oai_token-act-pair · gpt-3.5-turbo

phrases that express conditionality or hypothetical situations involving negation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.hook_resid_pre

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

\<

-0.70

è¦ļéĨĴ

-0.61

 Invisible

-0.60

ixed

-0.60

 Compass

-0.59

onut

-0.59

 simultane

-0.58

xxxx

-0.58

notation

-0.57

API

-0.56

POSITIVE LOGITS

 tolerate

0.81

 gladly

0.78

 prefer

0.78

 hesitate

0.78

 characterize

0.77

 classify

0.74

 recommend

0.74

 ideally

0.73

ĸļ

0.71

 dare

0.70

Activations Density 0.126%

contractions

phrases that express conditionality or hypothetical situations involving negation

No Comments

No Known Activations