INDEX

Explanations

the word "left" followed by various contextual settings or descriptions

oai_token-act-pair · gpt-3.5-turbo

references to the political spectrum, particularly the term "left" and related concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

cffff

-0.76

andise

-0.68

ãĤ©

-0.67

 Allaah

-0.66

riel

-0.66

antz

-0.66

glomer

-0.64

BuyableInstoreAndOnline

-0.63

ãĥīãĥ©

-0.63

idated

-0.62

POSITIVE LOGITS

overs

1.61

wing

1.08

ward

1.05

 flank

1.01

wich

1.00

 wing

0.98

over

0.97

 handed

0.95

 fielder

0.93

most

0.88

Activations Density 0.036%

the word "left" followed by various contextual settings or descriptions

references to the political spectrum, particularly the term "left" and related concepts

No Comments

No Known Activations