INDEX

Explanations

instances of phrases mentioning multiple items, often in a list or comparison

oai_token-act-pair · gpt-3.5-turbo

references to additional information or details

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ister

-0.85

ISTER

-0.72

 Patent

-0.70

ĸļ

-0.70

NRL

-0.69

ocene

-0.68

iker

-0.67

 Belfast

-0.66

onics

-0.64

 Bake

-0.64

POSITIVE LOGITS

 things

1.20

 factors

1.11

 feats

1.05

 considerations

1.00

 reasons

0.95

 topics

0.94

 places

0.87

 occasions

0.86

tions

0.86

things

0.85

Activations Density 0.041%

instances of phrases mentioning multiple items, often in a list or comparison

references to additional information or details

No Comments

No Known Activations