INDEX

Explanations

proper nouns related to different places, people, and organizations

oai_token-act-pair · gpt-3.5-turbo

references to the word "ip" 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

 EntityItem

-0.85

DEV

-0.70

 Dying

-0.69

 withd

-0.66

DERR

-0.65

 Faul

-0.61

ABS

-0.61

%]

-0.61

BIP

-0.61

 Baldwin

-0.60

POSITIVE LOGITS

artisan

1.42

otent

1.33

olitan

1.19

osit

1.19

arent

1.14

ublic

1.14

resent

1.11

rison

1.08

olit

1.08

olitics

1.08

Activations Density 0.043%

proper nouns related to different places, people, and organizations

references to the word "ip" in various contexts

No Comments

No Known Activations