INDEX

Explanations

phrases indicating familiarity or knowledge about a topic or subject

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "familiar" and its context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

emetery

-0.72

aughters

-0.69

alez

-0.65

 teasp

-0.61

rama

-0.61

estyles

-0.61

uld

-0.60

earchers

-0.59

verages

-0.59

 tribute

-0.58

POSITIVE LOGITS

 firsthand

0.80

è¦ļéĨĴ

0.68

 Norn

0.67

 whom

0.67

 regards

0.66

enough

0.66

 Divinity

0.65

 Plymouth

0.65

¥µ

0.64

 whats

0.63

Activations Density 0.112%

phrases indicating familiarity or knowledge about a topic or subject

occurrences of the word "familiar" and its context

No Comments

No Known Activations