INDEX

Explanations

words or phrases related to specific organizations or places marked by the letter 'N'

oai_token-act-pair · gpt-3.5-turbo

instances of the letter 'N'

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 Gibraltar

-0.73

 Sparrow

-0.69

espie

-0.68

 Cerberus

-0.67

gerald

-0.65

 Photographer

-0.65

 downed

-0.64

 unsupported

-0.63

 cant

-0.63

POSITIVE LOGITS

onsense

1.15

anny

1.14

usra

1.13

aughty

1.11

ucle

1.11

uggets

1.09

erves

1.07

ihil

1.07

omin

1.07

igg

1.06

Activations Density 0.040%

words or phrases related to specific organizations or places marked by the letter 'N'

instances of the letter 'N'

No Comments

No Known Activations