INDEX

Explanations

language related to digging or searching through something

oai_token-act-pair · gpt-3.5-turbo

the word "dig" and its various forms in the context of exploration or investigation

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

 Leilan

-0.77

 Blueprint

-0.74

BuyableInstoreAndOnline

-0.74

 wart

-0.68

 Karin

-0.63

 ACTIONS

-0.61

âĹ¼

-0.60

Thomas

-0.60

Thumbnail

-0.60

................................................................

-0.60

POSITIVE LOGITS

itial

1.06

gers

1.00

ongs

1.00

imon

0.98

raph

0.97

dug

0.95

dig

0.94

ression

0.92

Dug

0.91

ested

0.89

Activations Density 0.017%

language related to digging or searching through something

the word "dig" and its various forms in the context of exploration or investigation

No Comments

No Known Activations