INDEX

Explanations

actions or decisions being taken or made in various situations

oai_token-act-pair · gpt-3.5-turbo

phrases related to actions being taken or decisions made

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

 Shack

-0.62

waters

-0.60

 versatility

-0.59

 humble

-0.56

 itch

-0.55

 reinforcement

-0.55

inventoryQuantity

-0.54

pop

-0.54

rats

-0.54

inates

-0.54

POSITIVE LOGITS

 regarding

0.81

 whereby

0.76

.–

0.71

ALSE

0.71

by

0.69

},{"

0.69

.:

0.67

ocene

0.66

acial

0.65

 beforehand

0.65

Activations Density 0.218%

actions or decisions being taken or made in various situations

phrases related to actions being taken or decisions made

No Comments

No Known Activations