INDEX

Explanations

references to "Oz" or related terms such as "museum" and "Dorothy."

oai_token-act-pair · gpt-3.5-turbo

references to "Oz" and related cultural associations

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ãĥĩãĤ£

-0.83

ãĥł

-0.77

tenance

-0.72

ivities

-0.70

 Scots

-0.69

 à¨

-0.68

WAYS

-0.68

\">

-0.67

riott

-0.67

Chr

-0.66

POSITIVE LOGITS

aukee

1.05

Oz

1.04

pin

1.02

toc

0.96

eki

0.95

yssey

0.95

arro

0.95

opus

0.93

olit

0.93

acan

0.93

Activations Density 0.020%

references to "Oz" or related terms such as "museum" and "Dorothy."

references to "Oz" and related cultural associations

No Comments

No Known Activations