INDEX

Explanations

references or mentions of a specific entity named "Sec" with different numbers appended to it

oai_token-act-pair · gpt-3.5-turbo

sections regarding specific "Sec" classifications or designations, likely indicating a security or organizational context

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

lihood

-0.79

 Mata

-0.74

è¦ļéĨĴ

-0.70

 Warcraft

-0.67

 Manila

-0.66

Gat

-0.65

Interstitial

-0.64

 park

-0.64

FORE

-0.63

 Furious

-0.63

POSITIVE LOGITS

recy

1.10

ession

1.07

aucus

1.04

ular

0.99

unia

0.98

opian

0.89

rets

0.87

ularity

0.87

Sec

0.85

urities

0.85

Activations Density 0.013%

references or mentions of a specific entity named "Sec" with different numbers appended to it

sections regarding specific "Sec" classifications or designations, likely indicating a security or organizational context

No Comments

No Known Activations