INDEX

Explanations

phrases related to a particular standpoint or perspective

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate different viewpoints or perspectives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

tro

-0.78

und

-0.76

UES

-0.66

esc

-0.65

ilde

-0.65

lear

-0.65

weet

-0.64

ril

-0.64

ãĤĵ

-0.63

ĸļ

-0.63

POSITIVE LOGITS

 standpoint

0.97

 perspective

0.82

 merits

0.71

 perspectives

0.66

mma

0.66

 anyway

0.65

 consider

0.62

 Petr

0.62

 observers

0.61

 behalf

0.60

Activations Density 0.073%

phrases related to a particular standpoint or perspective

phrases that indicate different viewpoints or perspectives

No Comments

No Known Activations