INDEX

Explanations

phrases related to different roles or positions in organizations

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate roles or positions held by individuals in various organizations

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

illin

-0.78

llah

-0.77

edu

-0.75

mort

-0.71

lass

-0.70

BR

-0.69

Reward

-0.69

root

-0.68

plate

-0.68

ouse

-0.68

POSITIVE LOGITS

bidden

0.91

ked

0.90

geries

0.87

gotten

0.84

 example

0.83

 instance

0.82

gery

0.81

cers

0.81

aging

0.80

 Misc

0.79

Activations Density 0.109%

phrases related to different roles or positions in organizations

phrases that indicate roles or positions held by individuals in various organizations

No Comments

No Known Activations