INDEX

Explanations

phrases related to conflict, violence, and tragedy

oai_token-act-pair · gpt-3.5-turbo

references to significant historical events or incidents

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

 assistants

-0.63

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

-0.60

ãĤ´ãĥ³

-0.59

Ingredients

-0.58

 substitutes

-0.58

Accessory

-0.57

ITED

-0.57

OHN

-0.56

 antioxid

-0.55

aphael

-0.53

POSITIVE LOGITS

 massacre

0.77

 Massacre

0.73

waters

0.70

 Corridor

0.69

 Ruins

0.68

 Inferno

0.67

 corridor

0.67

 siege

0.67

 airst

0.66

 rivers

0.65

Activations Density 0.887%

phrases related to conflict, violence, and tragedy

references to significant historical events or incidents

No Comments

No Known Activations