INDEX

Explanations

actions related to destruction or modification

oai_token-act-pair · gpt-3.5-turbo

actions related to physical changes or disruptions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

soType

-0.72

Async

-0.70

else

-0.70

Calif

-0.68

Refer

-0.65

 EVENTS

-0.64

wcsstore

-0.64

ocating

-0.64

Asked

-0.63

arsity

-0.62

POSITIVE LOGITS

 several

1.22

 dozens

1.15

 numerous

1.13

 hundreds

1.13

 countless

1.10

 many

1.08

 innumerable

1.03

 thousands

1.02

 some

1.00

 tens

0.99

Activations Density 0.513%

actions related to destruction or modification

actions related to physical changes or disruptions

No Comments

No Known Activations