INDEX

Explanations

phrases related to cooperation or agreement

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate compliance or agreement

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

ilies

-0.80

ĳå£«

-0.74

ĻĤ

-0.72

tur

-0.72

orius

-0.72

ns

-0.71

iled

-0.67

orians

-0.65

aped

-0.65

iler

-0.65

POSITIVE LOGITS

ities

0.79

wagon

0.74

ments

0.73

leys

0.69

ity

0.69

side

0.68

 wagon

0.65

 along

0.63

 with

0.62

VERT

0.61

Activations Density 0.022%

phrases related to cooperation or agreement

phrases that indicate compliance or agreement

No Comments

No Known Activations