INDEX

Explanations

phrases related to security threats and concerns

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.8.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

-+-+

-0.70

acies

-0.69

ĻĤ

-0.66

ĳ

-0.65

©¶æ¥µ

-0.65

ãĥŃ

-0.65

ĪĴ

-0.65

ãĥł

-0.65

etsy

-0.64

Ń·

-0.63

POSITIVE LOGITS

selves

0.75

},"

0.73

,"

0.69

thumbnails

0.68

."[

0.66

".[

0.62

alone

0.61

.")

0.59

0.58

Â·Â·

0.58

Activations Density 0.041%

phrases related to security threats and concerns

No Comments

No Known Activations