INDEX

Explanations

names of people, locations, and organizations

oai_token-act-pair · gpt-3.5-turbo

specific names or terms related to notable individuals or organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

ahime

-0.81

orem

-0.77

erity

-0.65

REP

-0.61

vow

-0.58

itous

-0.58

aram

-0.58

 distance

-0.58

 miscarriage

-0.58

 obliged

-0.58

POSITIVE LOGITS

ĩ

0.76

 Kong

0.75

--+

0.74

ĵĺ

0.73

èª

0.73

ãĤ¡

0.68

Train

0.67

 Walton

0.66

«

0.66

PAC

0.65

Activations Density 0.286%

names of people, locations, and organizations

specific names or terms related to notable individuals or organizations

No Comments

No Known Activations