INDEX

Explanations

phrases related to levels of intensity

oai_token-act-pair · gpt-3.5-turbo

references to varying levels of intensity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

nee

-0.83

rooms

-0.76

vous

-0.75

rote

-0.71

OTA

-0.70

unes

-0.70

apan

-0.69

INTON

-0.69

arters

-0.68

 Passage

-0.67

POSITIVE LOGITS

 intensity

1.13

intensity

0.93

 è£ıè¦ļéĨĴ

0.88

 escalation

0.78

 intens

0.75

 intensify

0.74

uated

0.73

 levels

0.73

 level

0.72

ensive

0.72

Activations Density 0.019%

phrases related to levels of intensity

references to varying levels of intensity

No Comments

No Known Activations