INDEX

Explanations

words related to order or sequence

oai_token-act-pair · gpt-3.5-turbo

phrases related to procedures or steps to accomplish tasks

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ï¿½ï¿½ï¿½ï¿½

-0.69

wagen

-0.69

vous

-0.65

burgh

-0.65

rolet

-0.64

ãĤ©

-0.64

termination

-0.64

tek

-0.64

leigh

-0.63

tiny

-0.62

POSITIVE LOGITS

lies

0.95

liness

0.81

eering

0.80

 order

0.76

etary

0.75

eous

0.75

xual

0.74

ournals

0.73

 orders

0.71

ylum

0.70

Activations Density 0.027%

words related to order or sequence

phrases related to procedures or steps to accomplish tasks

No Comments

No Known Activations