INDEX

Explanations

phrases indicating contradiction or contrasting situations

oai_token-act-pair · gpt-3.5-turbo

terms related to contradictions and opposing viewpoints

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

î

-0.79

Animation

-0.68

inventoryQuantity

-0.65

Rocket

-0.65

anooga

-0.61

Down

-0.60

fell

-0.60

Family

-0.60

Same

-0.58

Image

-0.57

POSITIVE LOGITS

anwhile

0.57

 incidentally

0.55

 Ezek

0.55

 Lauder

0.54

 Kass

0.53

 Medals

0.53

 Berman

0.53

 Rudd

0.51

 Abbott

0.51

 Manz

0.50

Activations Density 1.223%

phrases indicating contradiction or contrasting situations

terms related to contradictions and opposing viewpoints

No Comments

No Known Activations