INDEX

Explanations

specific names of schools, places, or organizations

oai_token-act-pair · gpt-3.5-turbo

proper nouns and specific institutional names

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

rew

-0.86

paralle

-0.84

adal

-0.84

bre

-0.82

rez

-0.79

bat

-0.79

hex

-0.79

rongh

-0.78

tab

-0.78

sy

-0.77

POSITIVE LOGITS

 Networks

1.12

 Studios

1.10

 Publications

1.06

 Productions

1.05

 Industries

1.05

 Labs

1.03

 Corporation

1.03

 Airways

1.02

 Network

1.01

 Community

1.01

Activations Density 0.244%

specific names of schools, places, or organizations

proper nouns and specific institutional names

No Comments

No Known Activations