INDEX

Explanations

insights or revelations expressed in text

oai_token-act-pair · gpt-3.5-turbo

references to insights related to various topics

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

ategory

-0.70

senal

-0.65

 Vengeance

-0.62

 fuss

-0.60

HAM

-0.59

 Agric

-0.57

 sovere

-0.56

Ern

-0.56

 disbanded

-0.55

 Roses

-0.55

POSITIVE LOGITS

ibility

1.10

 glean

0.99

 insights

0.98

ively

0.98

 insight

0.93

ible

0.91

 into

0.88

ibly

0.87

spection

0.85

fully

0.82

Activations Density 0.058%

insights or revelations expressed in text

references to insights related to various topics

No Comments

No Known Activations