INDEX

Explanations

phrases related to permanence or remaining in a specific position or state

oai_token-act-pair · gpt-3.5-turbo

terms related to stability and continuity

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

apest

-0.77

iscons

-0.69

romy

-0.68

»Ĵ

-0.63

ĪĴ

-0.61

ĺħ

-0.60

eco

-0.59

ellar

-0.58

GI

-0.57

 sudden

-0.57

POSITIVE LOGITS

 indefinitely

1.65

 until

1.36

 till

1.29

 unchanged

1.29

until

1.22

 forever

1.19

 intact

1.15

 despite

1.13

til

1.12

withstanding

1.05

Activations Density 0.425%

phrases related to permanence or remaining in a specific position or state

terms related to stability and continuity

No Comments

No Known Activations