INDEX

Explanations

years mentioned in the text

oai_token-act-pair · gpt-3.5-turbo

references to specific time periods, particularly years and months

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

UTF

-0.72

BuyableInstoreAndOnline

-0.71

 kindred

-0.69

SOURCE

-0.63

uns

-0.62

Condition

-0.60

leeve

-0.59

utf

-0.59

 otherwise

-0.58

onne

-0.57

POSITIVE LOGITS

 however

0.78

 meanwhile

0.71

Ago

0.67

np

0.62

 æľ

0.62

 though

0.58

cific

0.57

ĩ

0.57

adays

0.56

eve

0.56

Activations Density 0.131%

years mentioned in the text

references to specific time periods, particularly years and months

No Comments

No Known Activations