INDEX

Explanations

the word "brunt."

oai_token-act-pair · gpt-3.5-turbo

references to the impact of negative events or situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

Liber

-0.79

Leader

-0.74

Sec

-0.72

croft

-0.71

Bound

-0.70

topia

-0.69

office

-0.68

hus

-0.67

Twe

-0.67

pocket

-0.66

POSITIVE LOGITS

 brunt

1.46

odox

0.87

 è£ıè¦ļéĨĴ

0.86

ï¸ı

0.84

icularly

0.81

 thereof

0.76

vre

0.75

cliffe

0.73

irez

0.72

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

0.69

Activations Density 0.006%

the word "brunt."

references to the impact of negative events or situations

No Comments

No Known Activations