INDEX

Explanations

prepositions followed by nouns or phrases

oai_token-act-pair · gpt-3.5-turbo

key terms related to conflict and scrutiny in a political context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 THEY

-0.61

they

-0.57

ANS

-0.56

They

-0.56

kaya

-0.52

 They

-0.51

ydia

-0.50

 yourselves

-0.50

 Thou

-0.50

ciation

-0.49

POSITIVE LOGITS

his

2.28

his

2.03

His

1.75

His

1.59

HIS

1.56

 himself

1.49

her

1.12

him

1.08

he

1.08

He

1.07

Activations Density 1.261%

prepositions followed by nouns or phrases

key terms related to conflict and scrutiny in a political context

No Comments

No Known Activations