INDEX

Explanations

references to occupying positions or places

oai_token-act-pair · gpt-3.5-turbo

the word "the" and variations of phrases involving notable roles or occupations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

fm

-0.67

idia

-0.66

ufact

-0.66

flix

-0.64

review

-0.64

ovych

-0.64

upt

-0.63

 disg

-0.63

 reports

-0.62

reflect

-0.61

POSITIVE LOGITS

ãĤ¦ãĤ¹

0.79

 imagination

0.74

 mythical

0.73

è¦ļéĨĴ

0.73

 Cosponsors

0.72

Merit

0.72

CHAT

0.70

 Territories

0.70

 territory

0.70

 mindset

0.70

Activations Density 0.342%

references to occupying positions or places

the word "the" and variations of phrases involving notable roles or occupations

No Comments

No Known Activations