INDEX

Explanations

phrases related to actions repeatedly done by individuals or groups

oai_token-act-pair · gpt-3.5-turbo

repeated mentions of the verb "have" in various contexts

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

wrap

-0.67

Reading

-0.65

 Apart

-0.62

Interested

-0.62

RIP

-0.60

 Isles

-0.59

 Meet

-0.59

 Finish

-0.59

iscovery

-0.58

opter

-0.57

POSITIVE LOGITS

 vowed

1.28

 been

1.22

 repeatedly

1.19

 insisted

1.14

 advocated

1.09

 argued

1.09

 resisted

1.07

 criticized

1.07

 struggled

1.07

 campaigned

1.06

Activations Density 0.185%

phrases related to actions repeatedly done by individuals or groups

repeated mentions of the verb "have" in various contexts

No Comments

No Known Activations