INDEX

Explanations

dates or time periods

oai_token-act-pair · gpt-3.5-turbo

specific years, particularly those related to significant events or articles

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

olar

-0.66

stand

-0.64

cko

-0.63

Dialogue

-0.63

edit

-0.62

urg

-0.62

task

-0.62

markets

-0.61

uana

-0.61

inqu

-0.60

POSITIVE LOGITS

ãĥ¼ãĥĨãĤ£

0.80

-'

0.73

abi

0.73

 maple

0.71

ãĥĥãĥī

0.70

 Coliseum

0.66

Fuk

0.66

 Cellular

0.65

ãĥŁ

0.65

é¾įå¥ĳå£«

0.64

Activations Density 0.165%

dates or time periods

specific years, particularly those related to significant events or articles

No Comments

No Known Activations