INDEX

Explanations

phrases related to statements and explanations made by individuals

oai_token-act-pair · gpt-3.5-turbo

statements attributed to individuals within a dialogue or reporting context

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

pee

-0.65

monary

-0.61

central

-0.60

VIS

-0.58

pd

-0.57

tail

-0.57

neau

-0.57

amac

-0.57

brance

-0.57

ible

-0.56

POSITIVE LOGITS

:]

0.83

weet

0.69

"[

0.67

aturdays

0.66

auga

0.66

:"

0.66

:'

0.64

 bluntly

0.63

 goodbye

0.62

utenberg

0.62

Activations Density 0.244%

phrases related to statements and explanations made by individuals

statements attributed to individuals within a dialogue or reporting context

No Comments

No Known Activations