INDEX

Explanations

numbers and measurements in the form of quantities

oai_token-act-pair · gpt-3.5-turbo

variations of the suffix "ort"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 bankruptcy

-0.65

ãĥ£

-0.63

 Chimera

-0.63

awa

-0.62

©¶æ

-0.61

-+-+

-0.60

pton

-0.59

ccording

-0.58

 Levant

-0.58

 Cage

-0.58

POSITIVE LOGITS

heast

1.40

heastern

1.32

gage

1.20

iculture

1.03

oise

1.03

icultural

1.02

une

0.98

raits

0.96

rait

0.95

iment

0.93

Activations Density 0.064%

numbers and measurements in the form of quantities

variations of the suffix "ort"

No Comments

No Known Activations