INDEX

Explanations

mentions of media organizations or news events

oai_token-act-pair · gpt-3.5-turbo

references to popular television shows or significant cultural events

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

 THEN

-0.55

 Intermediate

-0.55

"},{"

-0.54

ļéĨĴ

-0.52

then

-0.51

lication

-0.49

 laun

-0.49

atta

-0.49

apo

-0.49

OTAL

-0.48

POSITIVE LOGITS

 lately

1.35

 since

0.92

 steadily

0.87

since

0.82

 recent

0.80

Recent

0.69

 consistently

0.68

 recently

0.67

pmwiki

0.66

 countless

0.65

Activations Density 1.081%

mentions of media organizations or news events

references to popular television shows or significant cultural events

No Comments

No Known Activations