INDEX

Explanations

phrases related to consequences or outcomes

oai_token-act-pair · gpt-3.5-turbo

phrases indicating concern or implications for various groups and audiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

hess

-0.77

è¦ļéĨĴ

-0.77

qi

-0.72

ngth

-0.69

ohn

-0.67

æŃ¦

-0.67

aukee

-0.65

DAQ

-0.64

hao

-0.64

inv

-0.64

POSITIVE LOGITS

 anyone

1.11

us

1.07

 sure

1.06

 anybody

1.05

gotten

1.00

me

0.99

 everyone

0.98

 whoever

0.98

 everybody

0.95

 starters

0.94

Activations Density 0.204%

phrases related to consequences or outcomes

phrases indicating concern or implications for various groups and audiences

No Comments

No Known Activations