INDEX

Explanations

proper nouns related to politics and media figures

oai_token-act-pair · gpt-3.5-turbo

references to political figures, entities, and significant organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

range

-0.56

eworthy

-0.55

 Canaver

-0.54

margin

-0.53

inventoryQuantity

-0.51

erville

-0.51

umph

-0.49

ipeg

-0.49

 sqor

-0.49

ocamp

-0.49

POSITIVE LOGITS

 intervened

1.20

 interven

0.94

 couldn

0.92

 interfered

0.90

 gave

0.90

 opted

0.88

 forbid

0.86

 hadn

0.85

 took

0.84

 threw

0.84

Activations Density 0.758%

proper nouns related to politics and media figures

references to political figures, entities, and significant organizations

No Comments

No Known Activations