INDEX

Explanations

phrases related to decisions or actions taken by individuals or groups

oai_token-act-pair · gpt-3.5-turbo

references to significant decisions or actions that have important implications

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

 magazines

-0.75

models

-0.66

 watches

-0.63

 coats

-0.62

 inhabit

-0.61

 orbits

-0.60

 loves

-0.60

 Element

-0.60

 Candy

-0.59

 sizes

-0.59

POSITIVE LOGITS

 moot

0.92

 deterrence

0.82

AFTA

0.77

ĵĺ

0.75

 aimed

0.74

 vind

0.73

 jeopard

0.71

 imminent

0.70

 retrospective

0.70

 timed

0.69

Activations Density 0.591%

phrases related to decisions or actions taken by individuals or groups

references to significant decisions or actions that have important implications

No Comments

No Known Activations